AN ABSTRACT OF THE DISSERTATION OF
Jianyong Wu for the degree of Doctor of Philosophy in Chemistry presented on
November 23, 2009.
Title: Tandem Mass Spectrometric Analysis of Protein and Peptide Adducts of
Lipid Peroxidation-Derived Aldehydes
Abstract approved:
Dr. Claudia S. Maier
The adduction of proteins and other biomolecules by electrophilic lipid
peroxidation products such as 4-hydroxy-2-nonenal (HNE), 4-oxo-2-nonenal
(ONE), malondialdehyde (MDA) or acrolein (ACR) is thought to be an initiating
and/or propagating factor in the pathophysiology of several diseases such as
atherosclerosis, diabetes, Alzheimer’s, Parkinson’s and other age-related disorders.
The identification of protein sites modified by oxylipids is of key relevance for
advancing our understanding how oxidative damage affects structure and function
of proteins. Here, the use of MALDI tandem mass spectrometry with high energy
collision-induced dissociation (CID) on a TOF/TOF instrument for sequencing
oxylipid-peptide conjugates was systematically studied. Three synthesized model
peptides containing one nucleophilic residue (i.e. Cys, His or Lys) were reacted
with MDA, HNE, ONE and ACR. MALDI-MS analysis and MS/MS analysis
were performed to confirm the adduct type and the modification sites. Michael
adducts and Schiff bases were the predominant products under pH 7.4 within 2
hours. All MS/MS spectra of Michael adducts show the neutral loss of the
oxylipid moiety ions. MS/MS spectra of Cys-containing peptide oxylipid
conjugates exhibit additional characteristic neutral loss of HS-oxylipid moiety
ions. MS/MS spectra of His-containing peptide oxylipid conjugates show
characteristic oxylipid-containing His immonium ions. Spectra of Lys-containing
peptide oxylipid conjugates (Schiff base) also show oxylipid-containing Lys
immonium ions. However, there is no neutral loss of the oxylipid moiety ion for
these Schiff bases.
Determining the extent or relative amounts of the oxidative damage in cells could
provide valuable insights into the molecular mechanisms of the diseases caused
by oxidative stress. Relative quantitation of oxylipid-modified proteins in
biological samples is a challenging problem because of the complexity and
extreme dynamic range that characterize these samples. In this study, the reagents,
N'-aminooxymethylcarbonylhydrazino-D-biotin (ARP) and iodoacetyl-PEO2biotin (IPB), were used to enrich acrolein-modified Cys-containing peptides and
the corresponding unmodified ones from subsarcolemmal mitochondria (SSM).
The ratios between them were determined by nanoLC-SRM analysis. Model
Cys-containing peptides labeled with ARP-acrolein and IPB were employed to
demonstrate this method. Seven acrolein-modifed Cys-containing peptides from
five mitochondrial proteins were quantified. The ratios for those seven peptides
from the CCl4-treated rats are higher than the control ones indicating that the
ratios of acrolein-modified peptides to unmodified ones are potential markers of
oxidative stress in vivo.
Age-dependent changes of protein carbonyls were investigated in subsarcolemmal
mitochondria by using LC-SRM analysis of distinct ACR-modified Cys-containing
peptides. Immunochemical analysis using an anti-ACR monoclonal antibody
supported an increase of proteins modified by acrolein with age. However, total
protein carbonyls measurement using ARP in Western blot analysis did not
conform to this change suggesting that age-related changes in protein carbonyls
are complex and would benefit from more specific measurement protocols.
© Copyright by Jianyong Wu
November 23, 2009
All Rights Reserved
Tandem Mass Spectrometric Analysis of Protein and Peptide Adducts of Lipid
Peroxidation-Derived Aldehydes
by
Jianyong Wu
A DISSERTATION
submitted to
Oregon State University
in partial fulfillment of
the requirements for the
degree of
Doctor of Philosophy
Presented November 23, 2009
Commencement June 2010
Doctor of Philosophy dissertation of Jianyong Wu presented on November 23,
2009.
APPROVED:
Major Professor, representing Chemistry
Chair of the Department of Chemistry
Dean of the Graduate School
I understand that my dissertation will become part of the permanent collection of
Oregon State University libraries. My signature below authorizes release of my
dissertation to any reader upon request.
Jianyong Wu, Author
ACKNOWLEDGEMENTS
The author expresses heartfelt appreciation to his wife, Ping Shi, for her love,
support and understanding. The author is very appreciative for his major professor,
Dr. Claudia S. Maier for providing guidance, support, encouragement, and trust.
In addition, the author is grateful to Juan Chavez and the other members in Dr.
Maier’s group for support and encouragement. The author is also very thankful
for support from his committee members as well as Dr. Max Deinzer and Dr.
Staci Simonich. Many thanks are given to Michael Hare, Lilo Barofsky, Brian
Arbogast, Jeff Morre and the rest of the mass spectrometry group for their support.
Thanks are also given to Blaine R. Roberts for the help with peptide synthesis.
This work was supported by National Institutes of Health (NIH).
CONTRIBUTION OF AUTHORS
Dr. Claudia S. Maier provided leadership in all aspects of this dissertation. Juan
Chavez and the author wrote the Chapter 1 together as a mini review to be
submitted. Dr. Michael Hare provided suggestions in interpreting results and
edited chapter 3. Jie Zhang performed quantum calculation in Chapter 3. Kristina
Jonsson and Cristobal Miranda helped to perform some Western Blots in Chapter
5. The author helped with sample preparation and data analysis in the study
presented as Appendix A and B.
TABLE OF CONTENTS
Page
CHAPTER 1. Introduction.…………………….………….……………………. 1
CHAPTER 2. General Methods.…………………..………..………………….. 20
CHAPTER 3. Modification of Peptides by Lipid Peroxidation Products Studied by
MALDI-MS/MS using a TOF/TOF Instrument ……….……….. 45
Abstract.………………………………………………..……….. 47
Introduction.…………………………………………….………. 47
Material and methods……….…………………………….……. 49
Results and discussion………………………………………..… 52
Conclusions……………………………………………………... 57
Acknowledgements……………………………...…………….... 59
References…………………………………………………..…... 60
CHAPTER 4. Mass Spectrometry-based Quantification of Acrolein-modified
Cys-containing Peptides in an In Vivo Model of Oxidative
Stress .............................................................................................75
Abstract.…………………………………………………..…….. 76
Introduction.……………………….……………………………. 77
Material and methods……….…………….……………….……. 79
Results and discussion……………………..……………….…… 85
Conclusions……………………………………………………... 91
Acknowledgements………………………..……...…………….. 92
TABLE OF CONTENTS (Continued)
Page
References………………………………………………..……... 93
CHAPTER 5. Quantification of Peptide Adducts of Acrolein in Subsarcolemmal
Mitochondria in Aging Rats……………………..….……….....110
Abstract.………………………………………………..…..….. 111
Introduction.……………………………………….………..…. 111
Material and methods……….………………………….…..…. 113
Results and discussion…………………………………….…… 116
Conclusions…………………………………………………......118
Acknowledgements……………………………...…….….…….119
References………………………………………………..…..... 120
CHAPTER 6. CONCLUSIONS.…………………………….………………... 129
ABBREVIATION……………………………………………………..……..…131
BIBLIOGRAPHY…...……………………………………………….…..……. 134
APPENDICES ……………………………………………………..…….……. 143
Appendix A. A New Role for an Old Probe: Affinity Labeling of
Oxylipid Protein Conjugates by
N’-Aminooxymethylcarbonylhydrazino D-biotin ……….……. 143
Appendix
B.
Mitochondrial
Characterization
Proteins
by
of
Affinity
2-Alkenal
Labeling
Modified
and
Mass
Spectrometry…………………………………………………... 176
LIST OF FIGURES
Figure
Page
2.1. Schematic of the MALDI process ……………………………………..……37
2.2. Schematic of the ESI process under a positive mode ………………………37
2.3. Schematics of the three different mass spectrometers ……………..….……38
2.4. Two MS/MS modes often used for triple quadruples ………………………39
2.5. The nomenclature of peptide fragmentation ………………………….…….39
2.6. Common quantitative mass spectrometry workflows …………….…..…….40
2.7. Structure of the ICAT reagent …………….………………………….……..41
2.8. ITRAQ chemistry and application principle in mass spectrometry …….…..41
2.9. MALDI/MS spectra of the model peptide and its reaction products………..42
3.1. MALDI MS demonstrating adduction type of the Ac-SVVDLTHR-amide
peptide by HNE …………………………………………………….….…..68
3.2. MALDI MS/MS spectrum of HNE-peptide conjugate (Michael adduct)
Ac-SVVDLTCR-amide …………………………………………………….68
3.3. MALDI MS/MS spectrum of HNE-peptide conjugate (Michael adduct)
Ac-SVVDLTHR-amide ………………………………………………...….69
3.4. MALDI MS/MS spectrum of ONE-peptide conjugate (Michael adduct)
Ac-SVVDLTKR-amide ……………………………………………….……69
3.5. MALDI MS/MS spectrum of MDA-peptide conjugate (Schiff base)
Ac-SVVDLTKR-amide ……………………………………………....…….70
3.S-1-7. MALDI MS/MS spectra of Other oxylipid-peptide conjugates ………..71
4.1. Tandem mass spectra of the doubly charged Peptide a with different
modification …………………………………………………………..……99
LIST OF FIGURES (Continued)
Figure
Page
4.2. SRM chromatograms for the ARP-ACR modified model peptide and
IPB-labeled model peptide ………………………………………..…..…..100
4.3. Calibration curves for the ARP-ACR modified model peptide …………...101
4.4. SRM chromatograms for ARP-ACR modified peptides and PEO-labeled
peptides ……………………………………………………………..….…102
4.5. Ratios of acrolein-modified peptides to unmodified ones ………….......…102
4.S-1-14 Tandem mass spectra of the other doubly charged peptides with
different modifications …………………………………………..…103
5.1. Ratios of acrolein-modified peptides to unmodified ones for young rata and
old rats …………………………………………………………….............124
5.2. Images of SDS-PAGE and Western ARP blot of protein carbonyls for
mitochondrial proteins ……….………………………………….....……..125
5.3. Western blot analysis of protein carbonyls in SSM from young rats and old
rats ………………………………………………………………..……….126
5.4. Images of SDS-PAGE and Western blot of FDP-lysine Derivative for
mitochondrial proteins ……………………………………….…….……..127
5.5. Western blot analysis of FDP-lysine derivative in SSM from young rats and
old rats …………………………………..………………………..……….128
LIST OF TABLES
Table
Page
3.1. Calculated ELUMO and EHOMO for three lipid peroxidation products and amino
acids ……………………………………………………………….………66
3.2. Calculated nucleophilicity indexes (ω-) for lipid peroxidation product ……66
3.3. Peptide adduction type by lipid peroxidation products ………………….…67
3.4. Characteristic fragment ions for oxylipid-peptide conjugates ……….……..67
4.1. Summary of the seven acrolein-modified peptides ………………….…...…97
4.2. LC and MS properties for ARP-ACR and IPB labeled peptides …….….….97
4.3. Reproducibility test for our method ……………………………..……...…..98
5.1. Summary of the seven acrolein-modified peptides ………………...…..….123
LIST OF SCHEMES
Scheme
Page
1.1. Michael adducts formed from Cys-containing peptides ………….….….….19
2.1. Flow chart of mitochondrial isolation …………………………..….……….43
2.2. Flow chart of avidin affinity enrichment for biotin labeled peptides …….....43
3.1. Lipid peroxidation products generated from PUFA ………………..……….63
3.2. Michael adducts formed from Cys-containing peptides ……………………64
3.3. The aspartic acid effect ………………………………………………..……64
3.4. Pathways from Cys-containing peptide conjugate for neutral loss ……..….65
4.1. Structures and formations of ARP-ACR modified and IPB-labeled
peptides ………………………………………………………………...…..96
4.2. Relative quantitation procedure for acrolein-modified peptides ………...….96
5.1. Structures and formations of ARP-ACR modified and IPB-labeled
peptides ………………………………………………………………..….122
5.2. Relative quantitation procedure for acrolein-modified peptides in
mitochondrial protein samples……….........................................................122
Tandem Mass Spectrometric Analysis of Protein and Peptide Adducts of
Lipid Peroxidation-Derived Aldehydes
CHAPTER 1. Introduction
Mass Spectrometric Analysis of Oxidative Stress Induced Modifications to
the Proteome
Jianyong Wu and Juan Chavez
2
Introduction
Molecular oxygen (O2) is essential for the cellular respiration of all aerobic
organisms.
While its chemical reactivity provides a source of energy to drive the
process of life, it also makes O2 a potentially dangerous toxin.
Reactive oxygen
species (ROS) are highly reactive byproducts formed as a result of the use of
oxygen. A host of ROS including superoxide, hydrogen peroxide, hydroxyl radical,
peroxy radical and peroxynitrite are produced endogenously through numerous
metabolic pathways, including oxidative phosphorylation, oxidation of xanthine
by xanthine oxidase, cytochrome P-450 detoxification reaction and oxidation of
NADPH by NADPH oxidase [1-3]. Besides endogenous sources, ROS can also be
formed by exogenous agents like ozone, pesticides, photochemical smog and
ionizing radiation [3]. Although at high levels ROS are potentially damaging to
key cellular components, at lower levels they also play an important role in cell
signaling [4, 5]. Additionally they are very important in immune response, where
they are generated by neutrophils in lethal quantities to kill pathogens [6].
Conditions of oxidative stress can nonetheless result when cellular homeostasis is
disrupted due to excessive production of ROS and/or failure of the biological
system's ability to detoxify the reactive intermediates or repair the resulting
damage.
3
Mitochondria and Oxidative Stress
Serving as the primary energy plant for the cell, and consuming approximately
90% of the molecular oxygen taken in during aerobic respiration, mitochondria
are central to studies of oxidative stress. The mitochondrial electron transport
chain (ETC) complexes serve to reduce molecular oxygen into water, and in turn
generate a proton gradient across the inner mitochondrial membrane, which drives
the process of ATP production. Although electron transport through the complexes
is a highly efficient process, many studies have shown that a small but significant
percentage of the molecular oxygen consumed by the mitochondria, is nonetheless
converted into reactive oxygen species including the radical anion superoxide
(O.2-) and H2O2, due to electron “leakage” from the ETC complexes [7, 8].
Mitochondria are therefore a major source of superoxide within the cell [9]. On its
own, superoxide is relatively stable and innocuous, however it serves as a
precursor for other more reactive and damaging ROS including hydrogen
peroxide, peroxynitrite, and the hydroxyl radical [10]. Although there is some
debate about the actual mechanism of electron leakage, generally Complex I, and
Complex III are regarded as the primary production sites of ROS from the ETC.
In addition to the ETC, mitochondria contain a multiplicity of other
ROS-producing sources including cytochrome b5 reductase, monoamine oxidases,
dihydrooroate dehydrogenase, dehydrogenase of α-glycerophosphate, succinate
dehydrogenase, aconitase, and α-ketoglutarate dehydogenase.
4
However, the contribution to the total mitochondrial ROS production by these
seven sources remains unknown [11].
Antioxidant defense system
Under normal physiological conditions a balance of the ROS produced during
respiration is maintained by a complex multi-leveled antioxidant defense system
[11, 12]. This network of enzymes and non-enzymatic antioxidants serve to
protect proteins, DNA and other important biomolecules which are critical to the
function of the cell. The mitochondrial superoxide dismutase (Mn-SOD) serves
to eliminate superoxide by dismutation to form hydrogen peroxide, which can
then be catalytically decomposed into water and O2 through the action of
peroxidases. When the concentration of O.2- or H2O2 gets too high, H2O2 can
react with superoxide to form the extremely reactive hydroxyl radical (HO.).
Additionally H2O2 can undergo the Fenton reaction with divalent metal cations
such as Fe2+ and Cu2+ also producing HO..
The mitochondrial membrane, being rich in polyunsaturated fatty acids (PUFAs),
is particularly susceptible to ROS damage resulting in the lipid peroxidation
process. Typically antioxidants such as ubiquinione, ascorbic acid (vitamin C)
and the lipid soluble α-tocopherol (vitamin E), serve as radical scavengers thereby
protecting the vulnerable PUFAs. Additionally other small molecule antioxidants
including pyruvate, flavonoids, and carotenoids help to scavenge ROS [12]. One
5
of the most important and thoroughly studied mechanisms of antioxidant defense
is the glutathione-based detoxification system.
containing
tripeptide,
Consisting chiefly of thiol
glutathione (GSH, L-γ-glutamyl-L-cysteinylgylcine),
mitochondria utilize this system as both a renewable radical scavenger and as a
sacrificial consumable antioxidant [13]. Glutathione exists in a reduced (GSH)
and oxidized state (GSSG). In the reduced state, GSH can scavenge unstable
ROS including superoxide and hydroxyl radicals by functioning as an electron
donor to eliminate the reactive radicals. As a result GSH is converted into its
oxidized disulfide form GSSG.
The reduced form of GSH can be enzymatically
regenerated in the mitochondrial matrix by glutathione reductase. Additionally
in conjunction with the enzyme glutathione peroxidase, GSH can serve as an
electron donor in the conversion of hydrogen peroxide to water. Acting in a
sacrificial manner, the free thiol in GSH can react with electrophilic 2-alkenals
derived from the lipid peroxidation process, either non-enzymatically or in a
catalyzed
reaction
with
the
aid
of
glutathione-S-transferase.
The
glutathione-oxylipid conjugate can then be secreted from the mitochondria and the
cell [14]. Mitochondrial GSH is consumed in this process and therefore needs to
be replenished from the cytosolic GSH pool. Typically mitochondria maintain a
relatively high concentration of GSH. If a high enough concentration of GSH
cannot be maintained then oxidative stress induced damage will occur.
Under conditions of oxidative stress the antioxidant system becomes
6
overwhelmed and a host of lipid peroxidation products are formed.
Among
these, include the α，β-unsaturated aldehydes, such as HNE, ONE, HHE, acrolein
etc [15]. These lipid-derived electrophiles are known to be potent cellular toxins,
longer lived, and potentially more damaging to the cell than the ROS that
generated them [16].
They are capable of covalently modifying proteins at
nucleophilic sites including Cys, His and Lys through Michael-type addition or
Schiff base mechanisms, which can lead to enzymatic inhibition, increased ROS
production, and contribute to a cataclysmic cycle of oxidative stress induced
cellular damage. Scheme 1.1 shows examples of Michael adduct and Schiff base
between peptides and lipid peroxidation products.
Mitochondria are known to play a key role in the regulation of the cellular death
mechanisms of apoptosis and necrosis [17]. Damage induced by ROS and lipid
peroxidation products can lead to compromised mitochondrial integrity and the
release of proteins from the intermembrane space leading to cell death [18]. The
involvement of the mitochondrial permeability transition (MPT) in cell death is a
complex detailed subject which is outside the scope of this paper but has been
covered in many reviews [17, 19-22]. Basically oxidative stress can trigger MPT,
in which nonspecific pores are opened in the mitochondrial membrane leading to
a drop in the mitochondrial membrane potential (ΔΨ) and release of molecules
such as cytochrome C and Ca2+ that then activate programmed cell death pathways
[23]. MPT is known to play a role in several pathological conditions including
7
ischemia, stroke and myocardial infarction. The voltage dependent anion selective
channel (VDAC) and adenine nucleotide transporter (ANT) are both proteins
known to be important in the formation of MPT pores, and oxidative damage to
these proteins could play a role in the MPT process [24].
Protein Oxidation
The oxidative modification of proteins is a complex phenomenon, taking on many
forms and depending on a multitude of various factors.
The accumulation of
oxidized proteins has been associated with a number diseases as well as the
general aging process.
However the role of these modified proteins in the
etiology of disease is not well understood. Oxidative modification to proteins
can take many forms including radical induced cleavage of the peptide backbone,
direct oxidation of amino acid residue side chains, nitration and nitrosylation by
peroxynitrite, and conjugation with bioactive aldehydes such as lipid peroxidation
products.
The formation of protein carbonyl derivatives is widely used as a marker to assess
ROS induced damage during oxidative stress related to aging or age-related
diseases.
Protein carbonylation can occur as a result of direct oxidation of the
side chains of lysine, arginine, proline, and threonine by ROS such as HO.,
resulting in the formation of α-amino adipic semialdehyde for the case of lysine,
γ-glutamic
semialdehyde
for
the
case
of
proline
and
arginine,
and
8
aminoketobutyrate for the case of threonine.
Furthermore, carbonylated proteins
are formed by the reaction between the side chains of cysteine, histidine, and
lysine with α，β-unsaturated aldehydes such as HNE and acrolein.
The C3 atom
of α ， β-unsaturated aldehydes is electrophilic and readily reacts with the
nucleophilic sulfhydryl group of cysteine, the imidazole group of histidine and the
amino group of lysine, to form a Michael-type addition product.
Regardless of
how the carbonyl group is introduced, they exist as relatively reactive sites on the
protein molecule and can contribute to protein crosslinking by forming Schiff base
products with free amino groups in the same protein or on neighboring proteins.
Oxidatively crosslinked proteins are often enzymaticallly inactive and resistant to
proteolytic degradation [25-27]. It comes as no surprise that diseases associated
with an increase in protein carbonylation often also see an increase in protein
aggregate formation due to crosslinking.
Analytical methods for measuring protein carbonylation
Protein carbonylation is frequently used as a measure of oxidative stress induced
damage.
Therefore, a myriad of techniques have been developed for the analysis
of protein carbonyls. Some of the earliest developed, and most widely employed
methods rely on spectrophotometric analysis following derivatization with
thiobarbituric acid (TBARS) or 2, 4-dinitrophenylhydrazine (DNPH) [28, 29].
These methods are able to provide a measure of the global level of protein
9
carbonylation within a biological sample and are therefore useful in assessing the
degree of oxidative damage.
However they are not able to conclusively identify
the protein, specific amino acid site, or chemical nature of the oxidative
modification. Several methods have been developed to identify the carbonylated
proteins using the traditional proteomics analysis tools of gel electrophoresis and
Western blotting. Perhaps the most widely employed of these methods again
relies on the use DNPH as a labeling agent followed by immunochemical
detection with an anti-DNPH antibody [30]. Similar approaches have used biotin
hydrazide instead of DNPH, as a labeling agent, which can then be detected with
the use of avidin analogs and fluorescence or chemiluminescence detection [31,
32]. However, due to the lack of specificity of these hydrazine based reagents
towards the various types of carbonylation (e.g. 2-alkenal-protein adduct vs.
directly oxidized amino acid side chain), the chemical nature of the modification
remains unknown.
To this end several studies have employed immunological
detection using antibodies specific for the different types of alkenal-protein
adducts (i.e. anti-HNE or anti-acrolein antibodies) [33-36]. While these methods
can be highly specific for the differing forms of oxidative modification the
identity of the amino acid residue site of modification remains tentative at best.
Technological advances in mass spectrometry have provided researchers with
powerful tools to characterize post translational modifications (PTM) to proteins.
Mass spectrometry provides several advantages over other approaches for the
10
characterization of PTMs including the ability for unequivocal identification of
the amino acid site, chemical nature of PTM, discovery of novel PTMs, and the
ability to quantify relative changes in PTMs [37]. A variety of studies have
employed mass spectrometry to study protein carbonylation in model systems in
which purified proteins are incubated with LPO products or subjected to
metal-catalyzed oxidation reactions [38-44]. These studies are useful for
developing new analytical methods or for detailed structural and functional
analysis of proteins of interest, however the ultimate goal remains the analysis of
protein carbonyls in complex biological matrices. Several gel-free, mass
spectrometric methods have been developed in which researchers are able to
identify the protein target, amino acid site, and type of carbonylation in biological
samples.
Due to the low abundance of endogenously carbonylated proteins and
the extreme complexity of biological samples, affinity enrichment is necessary
prior to detection. Even still, many studies add exogenous agents to carbonylate
proteins in vitro before identifying the protein targets.
electrophile
probes
Using the biotin-tagged
N-iodoacetyl-N-biotinylhexylene-diamine
(IAB)
and
1-biotinamido-4-(4'-[maleimidoethylcyclohexane]-carboxamido) butane (BMCC)
researchers from the Liebler group were able to identify 897 different cysteine
residues from 539 proteins from nuclear and cytosolic fractions, and 1693
different cysteine residues from 809 proteins from the mitochondria [45, 46].
Several studies have employed the use of biotin-hydrazide derivatives as labeling
11
reagents for the enrichment and subsequent mass spectrometric analysis of protein
carbonyls [47-49]. Recently, our group has developed an analogous technique
employing N’-aminooxymethylcarbonylhydrazino-D-biotin (Aldehyde Reactive
Probe, ARP) to label, enrich and characterize oxidative modifications to proteins
[50]. This method represents an attractive alternative to hydrazine-based
derivatization methods for oxidized peptides and proteins because the reduction
step necessary for the transformation of the hydrazone bond to the chemically
more stable hydrazine bond can be omitted. Incubating mitochondrial proteins
with HNE followed by biotin hydrazide labeling, affinity enrichment and mass
spectrometric detection, Condreanu et al. were able to identify 18 specific sites of
HNE adduction. The flexibility of modern mass spectrometers allows for different
data-dependent acquisition methods combined with different dissociation
techniques including CID and ECD, to be applied for the identification of
carbonylated positions in peptides and proteins [38, 51-53]. Stevens el al.
identified 24 sites of Michael-type histidine-HNE adducts on 15 unique proteins
using neutral loss driven MS3 on a LTQ-FT mass spectrometer [53]. Employing a
solid phase hydrazide enrichment strategy and neutral loss driven MS3 analysis
Roe et al. mapped 125 sites of HNE adduction to 67 proteins from a yeast cell
lysate to which HNE had been added [52]. These in vitro studies demonstrate the
powerful potential of mass spectrometry in proteomics, and have proven useful
for identifying putative sites and protein targets of oxidative modification,
12
however it is still important to characterize the carbonyl modifications to proteins
occurring endogenously in complex biological systems. Applying our ARP based
analytical strategy to proteins isolated from rat cardiac mitochondria, we were
able to identify 39 unique sites on 27 proteins as being modified by a variety of
2-alkenal products including acrolein, β -hydroxyacrolein, crotonaldehyde,
4-hydroxy-2-hexenal, 4-hydroxy-2-nonenal and 4-oxo-2-nonenal (unpublished
results in Appendix B).
Quantitative analysis of carbonylated proteins based on mass spectrometry
Mass-spectrometry-based protein quantification has been developed greatly
recently. One major approach is based on isotope-labeled amino acids or peptides,
which can be recognized by a mass spectrometer according to their mass
difference [54-57]. Another major approach is label-free quantification including
two different strategies: (i) measuring and comparing signal intensity of peptides
from proteins; (ii) counting and comparing the number of identified-peptide
spectra from a particular protein [6, 58-62] (for an excellent review, see Ref [63] ).
All these approaches can be applied to quantification for carbonylated proteins as
long as the targeted peptides are replaced by carbonylatd ones. However low
abundance of carbonylated proteins in biological samples requires enrichment
methods not only for identification but also for quantification. Light and heavy
isotope labeled Girard’s P reagents, which can enrich carbonylated peptides by ion
13
exchange columns, were applied to establish the linear dynamic range for
quantification of a model carbonylated peptide by LC-MS [64].
Griffin’s group applied Multiplexed iTRAQ reagents to label digested peptides of
oxidized proteins from rat muscle mitochondria to establish the quantitative
reproducibility of the avidin enrichment strateg by LC-MS/MS [65]. Recently, our
group designed and synthesized an aldehyde-reactive, hydrazide-functionalized,
isotope-coded affinity tag (HICAT) and applied it to mitochondrial proteins
modified in vitro with HNE for quantitative analysis by nanoLC-MALDI-MS [53].
One acrolein modified peptide from ADP/ATP translocase from rat mitochondria
was also successfully quantified between young and old rats in vivo. A new
quantitative method without isotope-labeling is also being developed in our group.
The
reagent
N'-aminooxymethylcarbonylhydrazino-D-biotin
iodoacetyl-PEO2-Biotin
(IPB)
were
used
to
enrich
(ARP)
and
acrolein-modified
Cys-containing peptides and the corresponding unmodified ones from
mitochondrial samples. The ratios between them are determined by nanoLC-SRM
analysis using a hybrid linear ion trap mass spectrometer. While there have been
significant advances in the mass spectrometric identification and quantification of
carbonylated proteins, there is still much room for improvement on the current
methods.
More efficient and specific enrichment methods are critical since most
of quantitative MS analysis of endogenously carbonylated proteins only works
well for the most abundant of the low abundance target proteins, and there is
14
frequently significant interference from the very abundant non-modified proteins.
15
References
[1] M. Geiszt, J.B. Kopp, P. Varnai, T.L. Leto, Proc. Natl. Acad. Sci. U. S. A., 97
(2000) 8010.
[2] L.M. Sayre, M.A. Smith, G. Perry, Curr. Med. Chem., 8 (2001) 721.
[3] R.E. Pacifici, K.J. Davies, Gerontology, 37 (1991) 166.
[4] B. D'Autreaux, M.B. Toledano, Nat. Rev. Mol. Cell Biol., 8 (2007) 813.
[5] S. Nemoto, K. Takeda, Z.X. Yu, V.J. Ferrans, T. Finkel, Mol. Cell Biol., 20
(2000) 7311.
[6] L. Fialkow, Y. Wang, G.P. Downey, Free Radic. Biol. Med., 42 (2007) 153.
[7] A. Boveris, B. Chance, Biochem. J., 134 (1973) 707.
[8] J. St-Pierre, J.A. Buckingham, S.J. Roebuck, M.D. Brand, J. Biol. Chem., 277
(2002) 44784.
[9] A.J. Lambert, M.D. Brand, Methods Mol. Biol., 554 (2009) 165.
[10] J.S. Beckman, W.H. Koppenol, Am. J. Physiol., 271 (1996) C1424.
[11] A.Y. Andreyev, Y.E. Kushnareva, A.A. Starkov, Biochemistry (Mosc), 70
(2005) 200.
[12] T. Finkel, N.J. Holbrook, Nature, 408 (2000) 239.
[13] A. Pompella, A. Visvikis, A. Paolicchi, V. De Tata, A.F. Casini, Biochem.
Pharmacol., 66 (2003) 1499.
[14] J.D. Hayes, L.I. McLellan, Free Radic. Res., 31 (1999) 273.
[15] H. Esterbauer, R.J. Schaur, H. Zollner, Free Radic. Biol. Med., 11 (1991) 81.
[16] H. Esterbauer, Am. J. Clin. Nutr., 57 (1993) 779S.
[17] G. Kroemer, J.C. Reed, Nat. Med., 6 (2000) 513.
16
[18] S. Orrenius, V. Gogvadze, B. Zhivotovsky, Annu. Rev. Pharmacol. Toxicol.,
47 (2007) 143.
[19] A. Rasola, P. Bernardi, Apoptosis, 12 (2007) 815.
[20] Y. Tsujimoto, S. Shimizu, Apoptosis, 12 (2007) 835.
[21] S. Grimm, D. Brdiczka, Apoptosis, 12 (2007) 841.
[22] M. Crompton, Biochem. J., 341 ( Pt 2) (1999) 233.
[23] J.C. Yang, G.A. Cortopassi, Free Radic. Biol. Med., 24 (1998) 624.
[24] K.B. Choksi, W.H. Boylston, J.P. Rabek, W.R. Widger, J. Papaconstantinou,
Biochim. Biophys. Acta, 1688 (2004) 95.
[25] B. Friguet, FEBS Lett., 580 (2006) 2910.
[26] B. Friguet, L.I. Szweda, FEBS Lett., 405 (1997) 21.
[27] T. Grune, K.J. Davies, Mol. Aspects. Med., 24 (2003) 195.
[28] H. Ohkawa, N. Ohishi, K. Yagi, Anal. Biochem., 95 (1979) 351.
[29] R. Fields, H.B. Dixon, Biochem. J., 121 (1971) 587.
[30] A. Nakamura, S. Goto, J. Biochem., 119 (1996) 768.
[31] M. Oh-Ishi, T. Ueno, T. Maeda, Free Radic. Biol. Med., 34 (2003) 11.
[32] B.S. Yoo, F.E. Regnier, Electrophoresis, 25 (2004) 1334.
[33] D. Toroser, W.C. Orr, R.S. Sohal, Biochem. Biophys. Res. Commun., 363
(2007) 418.
[34] W.G. Chung, C.L. Miranda, C.S. Maier, Brain Res., 1176 (2007) 133.
[35] W.G. Chung, C.L. Miranda, J.F. Stevens, C.S. Maier, Food Chem. Toxicol., 47
(2009) 827.
[36]N. Tanaka, S. Tajima, A. Ishibashi, K. Uchida, T. Shigematsu, Arch. Dermatol.
Res., 293 (2001) 363.
17
[37] M.R. Larsen, M.B. Trelle, T.E. Thingholm, O.N. Jensen, Biotechniques, 40
(2006) 790.
[38] N. Rauniyar, S.M. Stevens, Jr., L. Prokai, Anal. Bioanal. Chem., 389 (2007)
1421.
[39] L.M. Sayre, D. Lin, Q. Yuan, X. Zhu, X. Tang, Drug Metab. Rev., 38 (2006)
651.
[40] M.S. Bolgar, S.J. Gaskell, Anal. Chem., 68 (1996) 2325.
[41] M. Carini, G. Aldini, R.M. Facino, Mass Spectrom. Rev., 23 (2004) 281.
[42] J.A. Doorn, D.R. Petersen, Chem. Res. Toxicol., 15 (2002) 1445.
[43] A.L. Isom, S. Barnes, L. Wilson, M. Kirk, L. Coward, V. Darley-Usmar, J.
Am. Soc. Mass Spectrom., 15 (2004) 1136.
[44] A. Temple, T.Y. Yen, S. Gronert, J. Am. Soc. Mass Spectrom., 17 (2006)
1172.
[45] M.K. Dennehy, K.A. Richards, G.R. Wernke, Y. Shyr, D.C. Liebler, Chem.
Res. Toxicol., 19 (2006) 20.
[46] H.L. Wong, D.C. Liebler, Chem. Res. Toxicol., 21 (2008) 796.
[47] H. Mirzaei, F. Regnier, Anal. Chem., 77 (2005) 2386.
[48] B.A. Soreghan, F. Yang, S.N. Thomas, J. Hsu, A.J. Yang, Pharm. Res., 20
(2003) 1713.
[49] P.A. Grimsrud, M.J. Picklo, Sr., T.J. Griffin, D.A. Bernlohr, Mol. Cell
Proteomics, 6 (2007) 624.
[50] J. Chavez, J. Wu, B. Han, W.G. Chung, C.S. Maier, Anal. Chem., 78 (2006)
6847.
[51] N. Rauniyar, S.M. Stevens, K. Prokai-Tatrai, L. Prokai, Anal. Chem., 81
(2009) 782.
[52] M.R. Roe, H. Xie, S. Bandhakavi, T.J. Griffin, Anal. Chem., 79 (2007) 3747.
18
[53] B. Han, J.F. Stevens, C.S. Maier, Anal. Chem., 79 (2007) 3342.
[54] S.P. Gygi, B. Rist, S.A. Gerber, F. Turecek, M.H. Gelb, R. Aebersold, Nat.
Biotechnol., 17 (1999) 994.
[55] A.J. Heck, J. Krijgsveld, Expert Rev. Proteomics, 1 (2004) 317.
[56] Y. Oda, K. Huang, F.R. Cross, D. Cowburn, B.T. Chait, Proc. Natl. Acad.
Sci.U. S. A., 96 (1999) 6591.
[57] K. Marley, D.T. Mooney, G. Clark-Scannell, T.T. Tong, J. Watson, T.M.
Hagen, J.F. Stevens, C.S. Maier, J. Proteome Res., 4 (2005) 1403.
[58] P.V. Bondarenko, D. Chelius, T.A. Shaler, Anal. Chem., 74 (2002) 4741.
[59] D. Bylund, R. Danielsson, G. Malmquist, K.E. Markides, J. Chromatogr. A,
961 (2002) 237.
[60] A. Gilchrist, C.E. Au, J. Hiding, A.W. Bell, J. Fernandez-Rodriguez, S.
Lesimple, H. Nagaya, L. Roy, S.J. Gosline, M. Hallett, J. Paiement, R.E. Kearney,
T. Nilsson, J.J. Bergeron, Cell, 127 (2006) 1265.
[61] D. Lin, H.G. Lee, Q. Liu, G. Perry, M.A. Smith, L.M. Sayre, Chem. Res.
Toxicol., 18 (2005) 1219.
[62] M.P. Washburn, D. Wolters, J.R. Yates, 3rd, Nat. Biotechnol., 19 (2001) 242.
[63] M. Bantscheff, M. Schirle, G. Sweetman, J. Rick, B. Kuster, Anal. Bioanal.
Chem., 389 (2007) 1017.
[64] H. Mirzaei, F. Regnier, J. Chromatogr. A, 1134 (2006) 122.
[65] D.L. Meany, H. Xie, L.V. Thompson, E.A. Arriaga, T.J. Griffin, Proteomics, 7
(2007) 1150.
19
Scheme 1.1. Michael adducts formed from Cys-containing peptides with ACR,
ONE and HNE (A), C-2 can also be the active site for ONE; Schiff bases formed
from Lys-containing peptides with MDA and ONE (B).
20
CHAPTER 2. General Methods
Mass Spectrometer
A mass spectrometer consists of an ion source, a mass analyzer and detector. A
sample is ionized in an ion source, and then the ions are sorted by an analyzer
according to their m/z values and detected by a detector. The introduction of
electrospray ionization (ESI) [1] and matrix-assisted laser/desorption ionization
(MALDI) [2] in 1980’s allows proteins and peptides, which are nonvolatile, to be
analyzed by MS. In MALDI technique, a matrix, like α-cyano-4-hydroxycinnamic
acid, is required. The matrix absorbs energy from the pulsed laser and transfers it
to the analyte, which is then desorbed and forms the protonated ion in gas phase
(Fig. 2.1). ESI is another popular soft ionization approach. High voltage (2-6 kV)
is applied between the emitter, where samples come from, and the inlet of the
mass spectrometer. The process of ESI involves the creation of an electrically
charged spray followed by the reduction of the droplets’ size and final liberation
of totally desolvated ions (Fig. 2.2).
Modern mass analyzers include Time-of Flight (TOF), Quadrupole (Q),
Quadrupole ion trap (QIT), Linear ion trap (LIT), Magnetic sector (B), Ion
cyclotron resonance (ICR) and Orbitrap. In this thesis, Time-of Flight (TOF),
Quadrupole (Q) and Linear Ion Trap (LIT) analyzers are employed. A quadrupole
analyzer uses oscillating electrical fields to selectively stabilize or destabilize the
paths of ions passing through a radio frequency (RF) quadrupole field [3]. This
21
analyzer can be operated under two modes: a RF-only mode or a RF/direct current
(DC) mode, which allows for m/z filtering. Under RF/DC mode, a stable
transmission of a narrow window of m/z can be acquired at a specified DC
amplitude. This high ion transmission feature makes quadrupole a popular choice
for precursor ion selection in tandem mass spectrometry. The RF-only mode is
often used to transfer and focus all ions. For a full MS scan in quadrupole, only
the picked ion can reach the detector and the others are wasted, thus the sensitivity
is poor. Recently, a few types of LIT were developed [4-7]. One of this techniques
commercialized by Applied Biosystems, Inc. is introduced here. Under this type
of LIT, a RF-only mode is used to hold ions radially and an entrance potential
barrier and an exit potential barrier are applied to contain ions axially. To eject the
ions to the detector, the drive and auxiliary RF voltages are simultaneously
ramped with an increase in amplitude, thus the ions with increasing m/z will
become unstable and be ejected for detection. The sensitivity in LIT is much
higher than quadrupole, since most of ions can reach the detector for full MS
scans.
The construction of a TOF analyzer was described in 1946 by Stephens [8]. Ions
with different m/z after moving through an acceleration voltage (U) are dispersed
in time (t) when they fly along a tube with a known length (l). Here is the equation
to describe this process: t = (m/2zeU)1/2l. According to this equation, flying time
is proportional to the square root of the m/z. Distribution of kinetic energy for
22
ions produced from an ion source makes the ions with the same m/z cannot reach
the detector at the exactly same time, thus reducing the resolution. One major
approach (Time-lag focusing) to reduce the distribution is to employ a delay
between generation and acceleration of the ions [9]. In this approach, ions are
generated in an environment without electric field and separated in space
according to their different initial velocities after a time delay. Then the
acceleration voltage is turned on with a fast pulse. The ions with higher initial
velocities have travelled further, thus will obtain less energy from the acceleration
voltage. Finally the spreads of kinetic energy for ions will be reduced. The other
approach is called reflector [10]. In this method, a retarding electric field, reflector,
is introduced in the ion passing way. The ions will penetrate the reflector and be
expelled in the opposite direction. The ions with higher kinetic energy will
penetrate deeper than those with the same m/z but having lower kinetic energy,
thus the ions with higher kinetic energy will have a longer journey to reach the
detector, achieving a correction in time-of-flight. TOF has become a high
resolution mass analyzer after these improvements.
In this thesis, there are three mass spectrometers used, which are 4000 Q-trap,
4700 TOF/TOF Proteomics Analyzer from Applied Biosystems and Q-TOF
Ultima Global from Waters. Fig. 2.3 shows schematics of these instruments.
Q-TOF Ultima Global in our lab operates in MS or MS/MS modes with a
nano-electrospray source. Under MS mode, the ions are generated in the ion
23
source and pass through the transfer optics, quadrupole (RF mode) and the
collision cell (no collision) to the pusher, where the ions will be accelerated and
pass the reflectron to reach the detector. Under MS/MS mode, the precursor ions
are selected by the quadrupole (RF/DC mode) and accelerated to the collision cell,
where collision with neutral gases happens. The fragment ions are then
accelerated and detected in a reflectron mode.
4000 Q-trap is a triple quadrupole linear ion trap mass spectrometer, in which the
second analyzer can function as quadrupole or LIT. LIT will be used to increase
sensitivity when normal MS or MS/MS is performed; quadrupole will be applied
when some special MS/MS for triple quadruple is required. In this thesis, the
precursor ion scan is used to find the precursor ions which contain specific
fragment ions after CID and the selected reaction monitoring (SRM) scan is
applied to quantitative analysis. In these two MS/MS modes, the second analyzer
is quadrupole. Under the precursor ion scan, the Q3 is fixed at a specified m/z and
Q1 sweeps a mass range, so the precursor ions containing that specified fragment
will be detected. Under the SRM scan, Q1 is fixed to at a specified m/z of M1 and
Q3 is fixed to F1, so only precursor ion with m/z of M1 and with fragment of F1
after CID can be detected by the detector. These two filter steps greatly reduce
interference making it a classic quantitative method for complex samples. SRM is
also a suitable mode to determine multiple analytes (like a few hundred analytes)
simultaneously because of the extremely fast switch between different SRM
24
transitions. Fig. 2.4 shows the schematic of these two MS/MS modes and one
SRM chromatogram of a peptide from a complex mitochondrial sample. The
relative low background indicates the specificity of SRM.
4700 TOF/TOF Proteomics Analyzer operates in MS or MS/MS modes with a
MALDI source. Under MS mode, generated ions are accelerated in Source 1 and
pass the tube in a linear way or a reflector way to reach the detector. Under
MS/MS mode, the precursor ions are chosen by the ion selector according the
flight time in a short linear TOF analyzer. After passing the collision cell, the
fragments are accelerated again in Source 2 and reach the reflector detector in the
linear way. Because of the shot first linear TOF tube, the resolution for
precursor-ion selection is usually only around 10 Da, which is much lower than
the previously introduced two mass spectrometers (about 1 Da) using a
quadrupole.
Proteomics based on mass spectrometry
Proteomics is the large-scale study of proteins, especially their structures and
functions [11, 12]. Mass spectrometry has become the most comprehensive and
versatile tool for proteomics study [13, 14]. Here, identification and quantification
of proteins based on mass spectrometry are mainly introduced. Right now,
bottom-up proteomics is the most popular method in which proteins are digested
into peptides prior to analysis of mass spectrometry. This approach usually
25
requires tandem spectra for peptide identification. Collision-induced dissociation
(CID) is the most prominent collision method although electron capture
dissociation (ECD) [15] and electron transfer dissociation (ETD) [16] become
useful especially for post-translational modifications recently. Under CID, the
isolated charged peptides (usually protonated) are activated by collisions with
inert gases, such as argon, helium or nitrogen, and are dissociated. The fragment
ions mainly are classified into two types. Ions containing N-terminus are labeled
with a, b and c; ions containing C-terminus are labeled with x, y and z. B and y
ions are the main fragment ions under CID although fragmentation patterns
change with other parameters, such as collision energy, amino acid composition
and pressure [17]. Fig. 2.5 shows the nomenclature of peptide fragmentation.
Tandem mass spectra of peptides are usually employed to identify the sequences
with software such as Mascot, SEQUEST, XProteo and Peaks, which have similar
principles. Here we focus on Mascot [18]. In this search engine, candidate
peptides based on different digestion enzymes and their theoretical tandem spectra
are generated in silico with protein databases and proprietary fragmentation
models. The experimental tandem mass spectra are then compared with theoretical
ones to calculate the probability that observed match between them is a chance
event. The match with the lowest probability is considered as the best match. A
widely accepted threshold is that the probability of the observed event occurring
by chance is lower than 5% (p < 0.05). For a good match, the probability (P) is
26
usually an extremely small number. A probability-based score, which is
-10Log10(P), is applied to give a convenient number. For example, the score is 60
if a probability is 10-6. A higher score means a higher match. During Mascot
searching, the identification score (p < 0.05) is important because it varies with
the probability and also the size of the data base used. Thus, Mascot scores alone
are not enough to show the quality of identification.
Quantification of peptides or proteins based on mass spectrometry mainly has two
approaches [19]. One major approach is based on isotope-labeled amino acids or
peptides, which includes reagents like ICAT or iTRAQ, methods like SILAC or
AQUA. The other major approach is label-free quantification including two
different strategies: (i) measuring and comparing signal intensity of peptides from
proteins; (ii) counting and comparing the number of identified-peptide spectra
from a particular protein. Fig. 2.6 shows these different ways, where boxes with
horizontal lines indicate when samples are combined. Earlier combination of
samples implies less variation involved during experiments, indicating more
accurate measurement. Label-free approaches are the least accurate for
quantification because all the systematic and non-systematic error happens
between experiments. Here we only introduce the isotope labeled approaches.
Stable isotope labeling by amino acids in cell culture (SILAC) [20] introduced by
Mann is the most popular metabolic labeling way, where the medium containing
13
C6-arginine and 13C6-lysine are used. In this way, the tryptic peptides can carry
27
at least one heavy-labeled amino acid to be distinguished with the light-labeled
ones. Usually, the MS intensities of heavy and light labeled peptides, which are
co-eluted from LC, are compared to measure their relative amount. Since the
samples are combined at the cell level, all quantitative error sources introduced by
the biochemical and mass spectrometric procedures are excluded, indicating
excellent accuracy. However, this method is mainly applied to immortalized cells
because of the high cost and long time required for creating heavy and light
labeling cells.
Chemical labeling of proteins or peptides is another way for quantification.
Theoretically, any reactive amino acid can be modified with isotope-labeled
chemicals [21]. Practically, side chains of lysine and cysteine are mainly applied
for this purpose. ICAT (Isotope-coded affinity tag) [22] and iTRAQ (Isotope tags
for relative and absolute quantification) [23] are the most popular reagents these
days. The ICAT reagent (Fig. 2.7) consists of three elements: an affinity tag
(biotin) that can enrich ICAT-labeled peptides by an avidin column; a linker,
which can carry light or heavy isotopes; and a reactive group with high specificity
to thiol groups (cysteines). Cysteines from samples react with light or heavy
ICAT, and the MS intensities of the same peptides from different samples will be
compared to quantify the two samples. Because cysteine residue is a relatively
rare in proteins, ICAT can significantly reduce the complexity of biological
samples, and meanwhile, it is not applicable to proteins which have no or few
28
cysteine residues.
ITRAQ reagents are specific to N-terminus of the peptides and ε-amino groups of
lysine residues via the specific N-hydroxysuccinimide (NHS) chemistry. Fig. 2.8
shows iTRAQ chemistry and application principle in mass spectrometry. The
iTRAQ consists of a reporter group, a mass balance group, and a peptide-reactive
group, which is specific with amine groups. The total masses for iTRAQ reagents
are the same and the masses for the reporter groups are different because of the
13
C, 15N and 18O atoms. The same peptides from different samples can be labeled
with iTRAQ containing different reporter groups. These labeled peptides have the
same precursor masses, while different fragment ions for reporter groups in
MS/MS. The relative concentration of the peptides can be measured from the
relative intensities of the corresponding reporter ions after CID. Since there are up
to eight different reporter groups available in iTRAQ today, this reagent is
qualified to quantify up to eight samples simultaneously, which is valuable for
biological samples with different ages.
Gerber et al. [24] developed a method called AQUA (absolute quantification of
protein) in 2004. In this approach, a known amount of synthesized isotope-labeled
peptide is added in a protein sample. After digestion, the relative intensity of this
peptide in heavy and light form is measured to figure out the absolute amount of
the protein containing this peptide. Unlike the other metabolic or chemical
labeling methods, where relative quantification is obtained for a large number of
29
proteins, AQUA method can obtain absolute quantification for one or a few
interested proteins depending on the number of the synthetic peptides added in a
sample. The other limitation for this approach is that the protein digestion
efficiency is assumed to 100 %, which is not the case for most biological samples.
Thus, Beynon et al. [25] refined this approach by creation of genes, which express
concatenated isotope-labeled peptides. These long peptides have close digestion
efficiency with proteins and multiple standard peptides after digestion can be
applied for multiple proteins of interest.
For all these quantitative approaches, MS usually is applied for the measurements
except the iTRAQ method, where MS/MS is used. One limitation for MS analysis
is the specificity since there usually are multiple isobaric peptides present in
complex samples. Another limitation is the dynamic range especially for a TOF
analyzer by using a micro-channel plate (MCP) detector with a time-to-digital
converter (TDC), which is easier to be saturated comparing with some other
detectors. Recently, SRM becomes popular for great reduction of these two
limitations [19, 26]. Two filter steps involved in SRM reduce interference and
only one fragment ion from the precursor ion is allowed to reach the detector,
reducing the chance of saturation.
Identification and quantification of carbonylated peptides were performed in this
thesis. An enrichment step was necessary before mass spectrometry analysis
because of their low abundances in mitochondrial samples. The reagents,
30
N'-aminooxymethylcarbonylhydrazino-D-biotin
iodoacetyl-PEO2-Biotin
(IPB)
were
used
(ARP)
to
label
and
acrolein-modified
Cys-containing peptides and the corresponding unmodified ones for enrichment.
The ratios between them are determined by nanoLC-SRM analysis.
Mitochondrial isolation
The experimental protocol for the animal studies was approved by the
Institutional Animal Care and Use Committee at Oregon State University. Male
F344 rats (Harlan, Indianapolis, IN) were housed in individual plastic cages
covered with Hepa filter and allowed free access to standard animal chow and
water ad libitum. Rats were anesthetized with ether and the hearts were perfused
with phosphate-buffered saline (PBS), removed, rinsed in saline, blotted dry, and
weighed. Subsarcolemmal (SSM) and interfibrillar (IFM) mitochondria were
isolated by the method used by Palmer et al. [27] with slight modification [28].
After trimming excess fat and removing the atria, the ventricles were minced and
homogenized 1:5 (w/v) with buffer A1 (200 mM mannitol, 5 mM MOPS, 2 mM
EDTA, 0.2 % BSA, pH 7.4). The heart tissues were homogenized with a glass
homogenizer and pestle attached to a drill. The homogenates were then
centrifuged at 500 × g for 10 min at 4°C (all remaining centrifugation steps were
also performed at 4°C). The supernatant was saved and the pellet was resuspended
in 10 volume of Buffer A1 and centrifuged at 500 × g for 10 min. The two
31
supernatants were subsequently pooled and centrifuged at 3000 × g for 10 min to
obtain SSM, while the pellet obtained in the second low-speed centrifugation step
was saved for IFM isolation. This pellet was resuspended in 3 volumes of Buffer
B1 (100 mM KCl, 50 mM MOPS, 2 mM EDTA, 0.2 % BSA, pH 7.4), nagarse (3
mg/g wet weight tissue) was added to the tissue pellet, incubated on ice for 0.5
min, and homogenized. The homogenate was then diluted 7-fold with Buffer B 1
and centrifuged at 5000 × g for 5 min. The supernatant was discarded and the
pellet resuspended in the original volume of buffer. The resuspended pellet was
centrifuged at 500 × g for 10 min to yield the nuclear pellet. The supernatant was
saved and the pellet was resuspended and centrifuged at 500 × g for 10 min. This
step was repeated once more with the supernatant being saved after each spin. The
supernatants from the three low-speed spins were combined and centrifuged at
3000 × g for 10 min to obtain the IFM pellet. The simple flow chart for this
procedure is listed in scheme 2.1.
Protein extraction from mitochondria
To obtain protein, the mitochondrial membranes needed to be broken. There were
two methods of protein extraction from mitochondria in this thesis. The
freeze/thaw method involved freezing mitochondria in a liquid nitrogen bath and
then thawing them at ice water with mild sonication. The second method involved
the melting the mitochondrial membranes by the 1% Triton - 100 detergent with
32
mild sonication at ice water. Finally, the proteins would be obtained by
centrifugation. The former method could extract 0.351 mg protein from two vials
of mitochondria; the latter one could obtain 0.585 mg. Detergent with mild
sonicatin had the higher efficiency and would be mainly used in this thesis.
Avidin affinity enrichment
Avidin is a tetrameric protein. Each identical subunit can bind to biotin with a
high degree of specificity and affinity. This avidin-biotin system can be applied to
enrich compounds with biotin tags. The avidin-biotin bond is one of the strongest
non-covalent bonds since the dissociation constant of avidin is about 10-15 M [29].
The native avidin has to be denatured to break the bond to release compounds
with biotin tags. This non-reversible nature of the avidin-biotin system can limit
its application in affinity enrichment. Thus, monomeric avidin with a reduced
affinity for biotin (KD ≈ 10-7 M) was developed [30]. This modified avidin allows
elution conditions from the avidin matrix much milder. In this thesis, monomeric
avidin is used and low pH conditions are used to release the target peptides with
biotin tags. Scheme 2.2 shows this enrichment procedure. The samples with biotin
labeled target peptides are loaded to the column with monomeric avidin beads
inside. Then the peptides without biotin tags are washed away with a buffer. The
final target peptides are eluted with a buffer of low pH.
33
Reaction properties of N'-aminooxymethylcarbonylhydrazino-D-biotin (ARP)
and iodoacetyl-PEO2-Biotin (IPB)
In this thesis, ARP and IPB were used to label the acrolein-modified peptides and
Cys-containing peptides, respectively. Scheme 2.3 showed the reaction to form
ARP-ACR modified and IPB-labeled peptides. The reaction yield and the stability
of the products were important for quantitative analysis. Here, one mM model
peptide (CLLLSAPRR) was applied to react with 2 mM IPB for one hour at room
temperature under darkness at pH 8.3. One mM this model peptide was also used
to react with acrolein to form the acrolein-modified peptide, which was further
used to react with 2 mM ARP at pH 7.4 to form the ARP-ACR modified peptide.
From the MALDI/MS spectra in Fig. 2.9, there was no native peptide signal for
the IPB labeling reaction and no acrolein-labeled peptide signal for ARP labeling
reaction, which indicates almost 100% reaction yield for these reactions. Three
mM DTT was added to IPB-labeled solution to remove excess IPB; and three mM
methanal was added to ARP labeling solution to remove excess ARP. These
solutions were then diluted by 100 times to close the real concentration in
biological samples. These solutions were reserved for 24 hours at pH 7.4 and 2
hours at pH 3.0 under room temperature, and then were determined by
MALDI/MS spectra (data not shown). These spectra were nearly identical to them
in Fig. 2.9. There was barely native peptide signal observed for the IPB labeling
solution or acrolein-labeled peptide signal for ARP labeling solution, which
34
indicated these products were stable enough for sample preparation and LC-MS
analysis.
35
References
[1] J.B. Fenn, M. Mann, C.K. Meng, S.F. Wong, C.M. Whitehouse, Science, 246
(1989) 64.
[2] M. Karas, F. Hillenkamp, Anal. Chem., 60 (1988) 2299.
[3] W. Paul, Angew. Chem. , 102 (1990) 780.
[4] B.A. Collings, W.R. Stott, F.A. Londry, J. Am. Soc. Mass Spectrom., 14
(2003) 622.
[5] J.W. Hager, Rapid Commun. Mass Spectrom. , 16 (2002) 512.
[6] Y. Huang, S. Guan, H.S. Kim, A.G. Marshall, Int. J. Mass Spectrom. Ion
Processes 152 (1996) 121.
[7] J.C. Schwartz, M.W. Senko, J.E. Syka, J. Am. Soc. Mass Spectrom., 13 (2002)
659.
[8] W.E. Stephens, Bull. Am. Phys. Soc., 21 (1946) 22.
[9] W.C. Wiley, I.H. McLaren, Rev. Sci. Instrum. , 26 (1955) 1150.
[10] B.A. Mamyrin, Int. J. Mass Spectrom. Ion Processes, 131 (1994) 1.
[11] N.L. Anderson, N.G. Anderson, Electrophoresis, 19 (1998) 1853.
[12] W.P. Blackstock, M.P. Weir, Trends Biotechnol., 17 (1999) 121.
[13] R. Aebersold, M. Mann, Nature, 422 (2003) 198.
[14] J.R. Yates, C.I. Ruse, A. Nakorchevsky, Annu. Rev. Biomed. Eng., 11 (2009)
49.
[15] R.A. Zubarev, Curr. Opin. Biotechnol., 15 (2004) 12.
[16] J.J. Coon, J.E. Syka, J. Shabanowitz, D.F. Hunt, Biotechniques, 38 (2005)
519.
[17] B. Paizs, S. Suhai, Mass Spectrom. Rev., 24 (2005) 508.
36
[18] D.N. Perkins, D.J. Pappin, D.M. Creasy, J.S. Cottrell, Electrophoresis, 20
(1999) 3551.
[19] M. Bantscheff, M. Schirle, G. Sweetman, J. Rick, B. Kuster, Anal. Bioanal.
Chem., 389 (2007) 1017.
[20] S.E. Ong, B. Blagoev, I. Kratchmarova, D.B. Kristensen, H. Steen, A. Pandey,
M. Mann, Mol. Cell Proteomics, 1 (2002) 376.
[21] S.E. Ong, M. Mann, Nat. Chem. Biol., 1 (2005) 252.
[22] S.P. Gygi, B. Rist, S.A. Gerber, F. Turecek, M.H. Gelb, R. Aebersold, Nat.
Biotechnol., 17 (1999) 994.
[23]P.L. Ross, Y.N. Huang, J.N. Marchese, B. Williamson, K. Parker, S. Hattan, N.
Khainovski, S. Pillai, S. Dey, S. Daniels, S. Purkayastha, P. Juhasz, S. Martin, M.
Bartlet-Jones, F. He, A. Jacobson, D.J. Pappin, Mol. Cell Proteomics, 3 (2004)
1154.
[24] S.A. Gerber, J. Rush, O. Stemman, M.W. Kirschner, S.P. Gygi, Proc. Natl.
Acad. Sci. U. S. A., 100 (2003) 6940.
[25] R.J. Beynon, M.K. Doherty, J.M. Pratt, S.J. Gaskell, Nat. Methods, 2 (2005)
587.
[26] J. Malmstrom, H. Lee, R. Aebersold, Curr. Opin. Biotechnol., 18 (2007) 378.
[27] J.W. Palmer, B. Tandler, C.L. Hoppel, J. Biol. Chem., 252 (1977) 8731.
[28] J.M. Duan, M. Karmazyn, Mol. Cell Biochem., 90 (1989) 47.
[29] N.M. Green, Biochem. J., 89 (1963) 585.
[30] R.A. Kohanski, M.D. Lane, Methods Enzymol., 184 (1990) 194.
37
Figure 2.1. Schematic of the MALDI process.
Figure 2.2. Schematic of the ESI process under a positive mode.
38
Figure 2.3 A. Schematic of 4000 Q-Trap (Picture modified from Applied
Biosystems, Inc. Concord, Ontario, Canada); B. Schematic of Q-TOF Ultima
Global (Picture modified from Micromass/ Waters, Manchester, UK); C.
Schematic of 4700 TOF/TOF Proteomics Analyzer (Picture modified from
Applied Biosystems, Inc. Framingham, MA).
39
Figure 2.4. Two MS/MS modes often used for triple quadruples: A, Precursor Ion
Scan; B, Selected Reaction Monitoring (SRM) Scan. C, one SRM chromatogram
of a peptide from a complex mitochondrial sample.
Figure 2.5. The nomenclature of peptide fragmentation. Y and b ions are the main
ions under CID.
40
Figure 2.6. Common quantitative mass spectrometry workflows. Boxes in red and
green represent two experimental conditions. Horizontal lines indicate when
samples are combined. Dashed lines indicate points at which experimental
variation and thus quantification errors can occur. (Figure is modified from
Bantscheff et al. [19].)
41
Figure 2.7. Structure of the ICAT reagent. The reagent exists in two forms: heavy
and light. (Figure is modified from Gygi et al. [22].)
Figure 2.8. ITRAQ chemistry and application principle in mass spectrometry. A,
Three parts for iTRAQ: a reporter group, a mass balance group (carbonyl), and a
peptide-reactive group (NHS ester). B, Structure after reaction between iTRAQ
and four peptides, the total masses are the same and the masses for the reporter
groups are different because of the 13C, 15N and 18O atoms. C, Four different
reporter groups under MS/MS are used to quantify the same peptide from four
different samples. (Figure is modified from Ross et al. [23].)
42
Figure 2.9 MALDI/MS spectra of the model native peptide (A); the reaction
products with IPB (B); the reaction products with acrolein and ARP (C).
43
Scheme 2.1. Flow chart of mitochondrial isolation.
Scheme 2.2. Flow chart of avidin affinity enrichment for biotin labeled peptides.
44
Scheme 2.3 The reactions to form ARP-ACR modified and IPB-labeled peptides.
45
CHAPTER 3.
Modification of Peptides by Lipid Peroxidation Products Studied by
MALDI-MS/MS using a TOF/TOF Instrument
Jianyong Wu, Jie Zhang, Michael Hare, Claudia S. Maier*
Department of Chemistry, Oregon State University, Corvallis OR, 97330
* Corresponding. author. Tel.: +1 541 737 9533; fax: +1 541 737 2062.
E-mail address: claudia.maier@oregonstate.edu (C. S. Maier).
46
Abstract
The adduction of proteins and biomolecules by 4-hydroxy-2-nonenal and/or other
lipid aldehydes is thought to be an initiating or propagating factor in the
pathophysiology of several diseases, such as athereosclerosis and diabetes, and
aging-related disorders e.g. Alzheimer’s disease (AD) and Parkinson’s disease.
The identification of protein sites modified by oxylipids is of key relevance for
advancing our understanding how oxidative damage affects structure and function
of proteins. Based on nucleophilicity indexes from quantum calculations and
percentages of reactive forms, the rank order of reactivity for amino acids at pH
7.4 is: Cys > His > Lys > Arg and the rank order of reactivity for lipid
peroxidation products is: 4-oxo-2-nonenal > acrolein > 4-hydroxy-2-nonenal.
Three synthesized model peptides containing one nucleophilic residue (i.e. Cys,
His or Lys) were reacted with malondialdehyde, 4-hydroxy-2-nonenal,
4-oxo-2-nonenal and acrolein. MALDI-MS analysis and MS/MS analysis under
high energy collision-induced dissociation (CID) were performed to confirm the
adduct type and the modification sites. Michael adducts and Schiff bases were the
predominant products under pH 7.4 within 2 hours. All MS/MS spectra of
Michael adducts contain the neutral loss of the oxylipid moiety ions.
Cys-containing peptide oxylipid conjugates have additional characteristic neutral
loss of HS-oxylipid moiety ions. His-containing peptide oxylipid conjugates have
characteristic oxylipid-containing His immonium ions. Lys-containing peptide
47
oxylipid conjugates (Schiff base) also show oxylipid-containing Lys immonium
ions. However, there is no neutral loss of the oxylipid moiety ion for these Schiff
bases.
Introduction
Reactive oxygen species (ROS), produced by various normal and abnormal
processes in the cell, can result in deleterious modification of proteins, DNA, and
lipids [1-5]. Oxidative stress due to ROS is counteracted by an intricate
antioxidant system, and the degree of oxidative stress is determined by the
balance of ROS production and antioxidant defenses [1]. Protein modifications
due to ROS have been implicated in age-related disorders and in the etiology and
pathology of inflammatory diseases. Oxidative modification of proteins can result
in both backbone peptide bond cleavage and side-chain modifications [3, 4, 6-8].
An important side-chain modification is the introduction of protein carbonyls by
secondary reaction of histidine (His), cysteine (Cys), lysine (Lys) and arginine
(Arg) residues with lipid peroxidation products, which are produced from
polyunsaturated fatty acids (PUFA) and their esters under oxidative stress [9-12].
Scheme 3.1 shows examples of lipid peroxidation products produced through
oxidized PUFA. The adduction of these lipid peroxidation products to proteins and
other biomolecules is thought to be an initiating or propagating factor in the
pathophysiology
of
several
diseases,
including
atherosclerosis,
cancer,
48
diabetes-related complications and neurodegenerative diseases [2, 13-18].
Malondialdehyde (MDA), 4-hydroxy-2-nonenal (HNE), 4-oxo-2-nonenal (ONE)
and acrolein (ACR) are four major lipid peroxidation products. The conjugation of
the double bond with the aldehyde group in HNE, ACR and ONE makes C-3 a
strong electrophilic center to form Michael adducts with nucleophilic sites in
proteins, such as histidine imidazole moieties, cysteine sulfhydryls and ε-amino
groups of lysine residues. C-2 can be another active site for Michael adducts in
ONE since the double bond is also conjugated with the ketone group [11]. In
addition to forming Michael adducts, these lipid peroxidation products are able to
form Schiff bases when the aldehyde group reacts with ε-amino groups of lysine
residues in proteins [2, 18-21]. Scheme 3.2 shows examples of Michael adduct
and Schiff base forming reactions of peptides with lipid peroxidation products.
The reaction rate of Michael addition of the nucleophilic side chains of Cys, His,
Lys, and Arg to ONE, HNE, and ACR have been determined experimentally
[21-26]. In addition, LoPachin et al. have used calculated quantum mechanical
parameters in conjunction with frontier molecular orbital (FMO) theory to
rationalize the observed reactivities [25-27]. In FMO theory, covalent bonding
occurs when an electron pair from the highest occupied molecular orbital (HOMO)
of the nucleophile is donated into the lowest unoccupied molecular orbital
(LUMO) of the electrophile. In a recently developed application of FMO theory,
calculated energies for the HOMO and LUMO are used to determine a
49
nucleophilicity index (ω-) which was shown to correlate with the relative reaction
rate for some nucleophilic reactions [28]. LoPachin et al. calculated nucleophilic
indexes for ACR and HNE reaction with unprotected Cys, His and Lys. In the
present study, nucleophilicity indexes for the reaction of protected Cys, His, Lys,
and Arg with ACR, HNE, and ONE have been calculated. Protected amino acids
were used for quantum calculations because the protected amino acid is expected
to better reflect its reactivity in a peptide.
Tandem mass spectrometry has emerged as the premier technology for the
identification of post-translational oxidative modification sites [29]. However, the
use of MALDI tandem mass spectrometry with high energy collision-induced
dissociation (CID) on a TOF/TOF instrument for sequencing oxylipid-peptide
conjugates has not been systematically studied. In this study, the adduction type of
oxylipid peptide conjugates was determined by MALDI-MS analysis. We also
systematically investigated the mass spectrometric fragmentation behavior of
synthetic oxylipid peptide conjugates containing different nucleophilic amino acid
residues modified by four different lipid peroxidation products using a TOF/TOF
instrument.
Material and methods
Chemicals
HNE (10 mg mL-1 in ethanol) and ONE (10 mg mL-1 in methyl acetate) were
50
obtained from Cayman Chemical, Ann Arbor, MI. 1,1,3,3-tetraethoxypropane and
acrolein were from Sigma, St. Louis, MO. 9-Fluoroenylmethoxy-carbonyl (Fmoc)
derivatized
amino
acids
were
from
SynPep
Co.
(Dublin,
CA).
Fmoc-Arg(pbf)-Rink Amide-MBHA resin with bound C-terminal residues was
purchased from AnaSpec Inc. (San Jose, CA).
Synthesis of MDA and modified peptides.
The
peptides
Ac-SVVDLTCR-amide,
Ac-SVVDLTHR-amide,
Ac-SVVDLTKR-amide were synthesized using a PS3 automated solid phase
peptide synthesizer (Protein Technologies, Inc., Tucson, AZ). MDA was
synthesized by reacting 1,1,3,3-tetraethoxypropane with HCl according to the
procedures in Guy et al. [30]. The MDA solution was neutralized with NaOH
before reacting with peptides. To synthesize oxylipid-peptide conjugates, 1 µmol
peptide in 1 mL 10 mM tricine buffer, pH 7.4, was reacted with 10 µmol lipid
peroxidation products at 37 oC for two hours. Oxylipid peptide conjugates were
analyzed by MALDI-MS and MALDI-MS/MS.
Mass spectrometry
MALDI-MS and MALDI-MS/MS analysis was performed on an ABI 4700
Proteomics Analyzer with TOF/TOF optics equipped with a 200-Hz
frequency-tripled Nd:YAG laser operating at a wavelength of 355 nm (Applied
51
Biosystems, Inc. Framingham, MA). The reflector mode was used for all peptide
mass spectra. The accelerating voltage was set to 20 kV for MS, and 8 kV for
MS/MS. For MS/MS experiments, the collision energy was 1 kV, defined by the
potential difference between the source acceleration voltage (8 kV) and the
floating collision cell (7 kV). The precursor ion was selected by operating the
timed gate window with a nominal resolution of 50 ppm. Gas pressure (air) in the
collision cell was set to 6 x 10-7 Torr. Fragment ions were accelerated by 15 kV
into the reflector. Mass data were acquired in a mass range of 700-4000 Th with
external calibration using AB’s 4700 calibration mixture consisting of the
following
peptides:
des-Arg1-bradykinin
([M+H]+
at
m/zcalc
904.4675),
angiotensin 1 ([M+H]+ at m/zcalc 1296.6847, Glu1-fibrinopeptide B ([M+H]+ at
m/zcalc 1570.6768), and ACTH 18-39 ([M+H]+ at m/zcalc 2465.1983). The MALDI
matrix was α-cyano-4-hydroxycinnamic acid (Pierce, Rockford, IL).
Quantum mechanical calculations
Lowest unoccupied molecular orbital (LUMO) and highest occupied molecular
orbital (HOMO) energies (ELUMO and EHOMO) were calculated using Gaussian 03
software [31]. The molecular geometries were optimized with a 6-31G* basis set
and single-point energies were calculated using density functional theory (DFT)
with a B3LYP-6-31G* basis set. Hardness (η) was calculated as η = (ELUMO EHOMO)/2 and chemical potential (µ) was calculated as µ = (ELUMO + EHOMO)/2.
52
The nucleophilicity index (ω-) was calculated as ω- = ηA(µA-µB)2/2(ηA+ηB)2, where
A is the reacting nucleophile and B is the reacting electrophile. Higher
nucleophilicity indexes are correlated with higher reaction rates between
nucleophiles and electrophiles [28].
Results and discussion
Adduction of peptides by lipid peroxidation products
The three model peptides Ac-SVVDLTCR-amide, Ac-SVVDLTHR-amide, and
Ac-SVVDLTKR-amide were chosen as representative of the tryptic peptides
typically encountered in proteomics. Arginine was selected as the less reactive of
the two possible tryptic C-terminal residues. The N-terminus of the three model
peptides was acetylated to prevent possible interferences with the ε-amino group
of lysine and the imidazole of histidine. Arginine was selected as the C-terminal
residue for the model peptides to mimic tryptic peptides, which were widely
produced in proteomics study. The adduction type of oxylipid peptide conjugates
was determined by the differing mass increment for Michael adduct and Schiff
base formation. As shown in Fig. 3.1, the Michael adduct, with a mass increment
of 156.1 Da, is the predominant product of the reaction of the His-containing
peptide with HNE. Cys-containing and Lys-containing peptides also form Michael
adducts with HNE (MS spectra not shown). Arg has not been observed to react
with HNE. These results are consistent with our calculations and experimental
53
results of Doorn et al. [23].
The more electrophilic compounds ONE and ACR form Michael adducts with all
three model peptides. ONE and ACR have 154.1 Da and 56.0 Da adduct mass
increments, respectively (MS spectra not shown). There is no Michael adduction
observed between MDA and peptides, since MDA exists primarily as the less
electrophilic enolate anion (-O-CH=CH-CH=O) at pH 7.4. However, MDA and
ONE form Schiff bases with the Lys-containing peptide, and 54.0 Da and 136.1
Da adduct mass increments for MDA and ONE respectively are observed (MS
spectra not shown). The adduction products are summarized in Table 3.3.
In
this chapter we have focused on the Michael adducts and Schiff bases. Other
products are known to be produced after longer reaction times with higher
concentration of oxylipid [32-34].
Reactivity of lipid peroxidation products and amino acids to form adducts
To form adducts, ONE, HNE and ACR are electrophiles and deprotonated Cys,
neutral His, Lys, Arg are nucleophiles although only some of them exit as these
reactive forms at pH 7.4 according to their pKa values from side chains.
Acetylation of N-termini and amidation of the C-termini of these amino acids
mimic peptide bonds and prevent interferences from these available lone electron
pairs as nucleophilic centers. Table 3.2 shows nucleophilicity indexes (ω-) for
lipid peroxidation product reactions with possible nucleophilic targets using
54
orbital energies of ELUMO and EHOMO from Table 3.1. Nucleophilicity indexes for
ONE with nucleophilic targets are higher than those for ACR and HNE indicating
that ONE is the most reactive. ACR and HNE have similar nucleophilic indexes,
however, HNE has a slower rate of Michael addition due to the steric effect of the
alkane tail as discussed by LoPachin et al. [35]. Table 3.2 also shows that
nucleophilic indexes with Cys are the highest and those with His, Lys and Arg are
very close and much lower than Cys, which makes the Cys the most reactive
although the deprotonated Cys accounts for about 10% according to its pKa value
of 8.2. Because of the side chain pKa values for His, Lys and Arg (His 6.0; Lys
10.5; Arg 12.5), about 90% of His, 0.1 % of Lys and 0.001% of Arg exit as neutral
forms at pH 7.4, which are nucleophilic. Considering the similarity of the
nucleophilic indexes and the huge percentage differences of reactive forms for His,
Lys and Arg, His is the most reactive and Arg is the least. Higher pH will increase
percentage of neutral forms of Lys and Arg and increase their reactivity. Thus, the
rank order of reactivity for these four amino acids at pH 7.4 is: Cys > His > Lys >
Arg. The reactivity predicted mainly by the quantum mechanical calculations and
the percentages of reactive forms under pH 7.4 are agreement with those based on
reaction rates induced from experiments [21-24].
55
Characteristics of MALDI MS/MS spectra of oxylipid-modified peptides
Oxylipid-modified Cys-containing peptides
The MALDI MS/MS spectra of oxylipid peptide conjugates in our experiments
show that the oxylipid can be well retained with many fragments, which is a
prerequisite for peptide identification based on the use of uninterpreted tandem
mass spectra acquired in conjugation with automated database searching. For
example, the MS/MS spectrum of the HNE Cys-containing peptide conjugate in
Fig. 3.2 shows the first C-terminal y1 ion and fragment ions, y2*-y5*, which retain
the HNE moiety. Obviously, these peptide fragment ions confirm that the Cys
residue is modified by HNE. The strong y4* fragment ion is observed because of
the presence of Asp residue in the peptide. Scheme 3.3 shows one explanation by
charge-remote peptide fragmentation pathway (Paizs et al. [36]). This spectrum
exhibits two characteristic fragment ions at 933.4 Th and 899.4 Th, which
indicates neutral loss of HNE moiety and HS-HNE (C9H18O2S) moiety from the
conjugate, respectively. The possible pathways are based on a reversal Michael
adduction mechanism and a β-elimination mechanism, which are shown in
Scheme 3.4. MS/MS spectra of the other two oxylipid-modified Cys-containing
peptides show ions with oxylipid and HS-oxylipid neutral losses too (Fig. 3.S-1,
2). These neutral losses make the identification of oxylipid-modified peptides
more challenging. However, this feature can also be used to indentify
HNE-modified peptides with neutral loss-driven MS3 technique by Roe et al. [37]
56
and Rauniyar et al. [38].
Oxylipid-modified His-containing peptides
The MS/MS spectrum of the HNE His-containing peptide conjugate in Fig. 3.3
shows two characteristic fragment ions at 967.6 Th and 266.2 Th, which are the
neutral loss of HNE moiety from the conjugate and the HNE-containing His
immonium ion, respectively. Spectra for ACR and ONE His-containing peptide
conjugates
also
exhibit
neutral
loss
of
oxylipid
moiety
ions
and
oxylipid-containing His immonium ions, which are 166.1 Th and 264.2 Th,
respectively (Fig. 3.S-3, 4). The pathway for neutral losses of oxylipid moieties
from the conjugates is the same with the one in Scheme 3.3 since they have the
–CH2CH2CH=O structure in common. MALDI-TOF/TOF spectra did not show
the protonated dehydrated HNE moiety (139.1 Th), which commonly appeared in
low energy-CID tandem mass spectra [37, 39, 40].
Oxylipid-modified Lys-containing peptides
The MS/MS spectrum of the ONE Lys-containing peptide conjugate (Michael
adduct) in Fig. 3.4 shows a neutral loss of ONE moiety ion from the conjugate at
958.6 Th, Lys immonium ion with neutral loss of NH3 at 84.1 Th and
ONE-containing Lys immonium ion with neutral loss of NH3 at 238.2 Th. The
MS/MS spectra of the ACR and HNE Lys-containing peptide conjugates (Michael
57
adduct) indicate the same characteristic fragment ions showing the ACR/HNEcontaining Lys immonium ion with neutral loss of NH3 at 140.1/240.2 Th (Fig.
3.S-5,6). The pathway for neutral losses of oxylipid moieties from the conjugates
is the same with the one in Scheme 3.3 when the S element in side chain of Cys
was replaced by the N element in Lys.
The MS/MS spectrum of the MDA Lys-containing peptide conjugate (Schiff base)
in Fig. 3.5 only shows one strong characteristic fragment ion at 138.1 Th, which is
MDA-containing Lys immonium ion with neutral loss of NH3. There was no
neutral loss of MDA moiety or Lys immonium related ions from the conjugate.
The MS/MS spectrum of the ONE Lys-containing peptide conjugate (Schiff base)
also only shows one ONE-containing Lys immonium characteristic ion at 237.2
Th and no neutral loss of ONE moiety ion (Fig. 3.S-7). These results imply there
is no reasonable pathway to remove the oxylipd moiety part from Schiff bases.
These MS/MS spectrum differences between Michael adducts and Schiff bases
can be used to determine the adduction type of peptides modified by oxylipids
under high energy-CID. However, under low energy-CID, Rauniyar et al.
observed ions resulting from the neutral loss of HNE moiety from an
HNE–modified Lys-containing peptide (Schiff base) with a linear ion trap [41].
Conclusions
Based on nucleophilic indexes from quantum calculations and percentages of
58
reactive forms in solution, the rank order of reactivity for amino acids at pH 7.4 is:
Cys > His > Lys > Arg. The rank order of reactivity for lipid peroxidation
products at pH 7.4 is: ONE > ACR > HNE. In our study, the MS/MS spectra allow
obtaining sequence information for oxylipid-peptide conjugates which, at least in
favorable cases, enables unambiguous localization of the modification sites by
lipid peroxidation products. Compared with tandem mass spectrum under low
energy-CID, no protonated dehydrated HNE moiety (139.1 Th) is observed for
His-containing peptide oxylipid conjugates under high energy-CID. All the
MS/MS spectra of Michael adducts show the neutral loss of the oxylipid moiety
ions because they have –CH2CH2CH=O structure in common. Spectra of
Cys-containing peptide conjugates exhibit additional characteristic neutral loss of
HS-oxylipid moiety ions. In addition to neutral loss of oxylipid moiety ions,
spectra of His/Lys-containing peptide conjugates (Michael adduct) show
characteristic oxylipid-containing His/Lys immonium or related ions. Spectra of
Lys-containing peptide conjugates (Michael adduct) show an additional Lys
immonium ion with neutral loss of NH3. Spectra of Lys-containing peptide
conjugates (Schiff base) also show oxylipid-containing Lys immonium ions or
ions with neutral loss of NH3. All characteristic fragment ions are summarized in
Table 3.4.
59
Acknowledgements
This work was supported by grants from the NIH/NIA (AG025372). The mass
spectrometry core facility of the Environmental Health Sciences Center at Oregon
State University is supported in part by a grant from NIEHS (ES00210).
60
References
[1] T. Finkel, N.J. Holbrook, Nature, 408 (2000) 239.
[2] L.J. Marnett, J.N. Riggins, J.D. West, J. Clin. Invest., 111 (2003) 583.
[3] R.L. Levine, E.R. Stadtman, Exp. Gerontol., 36 (2001) 1495.
[4] E.R. Stadtman, Ann. N. Y. Acad. Sci., 928 (2001) 22.
[5] J.D. West, L.J. Marnett, Chem. Res. Toxicol., 19 (2006) 173.
[6] R.T. Dean, S. Fu, R. Stocker, M.J. Davies, Biochem. J., 324 ( Pt 1) (1997) 1.
[7] R.L. Levine, Free Radic. Biol. Med., 32 (2002) 790.
[8] B.S. Berlett, E.R. Stadtman, J. Biol. Chem., 272 (1997) 20313.
[9] H. Esterbauer, M. Dieber-Rotheneder, G. Waeg, G. Striegl, G. Jurgens, Chem.
Res. Toxicol., 3 (1990) 77.
[10] H. Esterbauer, R.J. Schaur, H. Zollner, Free Radic. Biol. Med., 11 (1991) 81.
[11] L.M. Sayre, D. Lin, Q. Yuan, X. Zhu, X. Tang, Drug Metab. Rev., 38 (2006)
651.
[12] C. Schneider, K.A. Tallman, N.A. Porter, A.R. Brash, J. Biol. Chem., 276
(2001) 20831.
[13] J.M. Gutteridge, B. Halliwell, Trends Biochem. Sci., 15 (1990) 129.
[14] R. Bucala, A. Cerami, Adv. Pharmacol., 23 (1992) 1.
[15] G. Jurgens, J. Lang, H. Esterbauer, Biochim. Biophys. Acta, 875 (1986) 103.
[16]W. Palinski, M.E. Rosenfeld, S. Yla-Herttuala, G.C. Gurtner, S.S. Socher, S.W.
Butler, S. Parthasarathy, T.E. Carew, D. Steinberg, J.L. Witztum, Proc. Natl. Acad.
Sci. U. S. A., 86 (1989) 1372.
[17] N. Traverso, S. Menini, L. Cosso, P. Odetti, E. Albano, M.A. Pronzato, U.M.
Marinari, Diabetologia, 41 (1998) 265.
61
[18] K. Uchida, Free Radic. Biol. Med., 28 (2000) 1685.
[19] J.R. Requena, M.X. Fu, M.U. Ahmed, A.J. Jenkins, T.J. Lyons, J.W. Baynes,
S.R. Thorpe, Biochem. J., 322 ( Pt 1) (1997) 317.
[20] L.M. Sayre, M.A. Smith, G. Perry, Curr. Med. Chem., 8 (2001) 721.
[21] D. Lin, H.G. Lee, Q. Liu, G. Perry, M.A. Smith, L.M. Sayre, Chem. Res.
Toxicol., 18 (2005) 1219.
[22] J.A. Doorn, S.K. Srivastava, D.R. Petersen, Chem. Res. Toxicol., 16 (2003)
1418.
[23] J.A. Doorn, D.R. Petersen, Chem. Res. Toxicol., 15 (2002) 1445.
[24] X. Zhu, L.M. Sayre, Chem. Res. Toxicol., 20 (2007) 165.
[25] R.M. LoPachin, D.S. Barber, T. Gavin, Toxicol. Sci., 104 (2008) 235.
[26] R.M. LoPachin, T. Gavin, B.C. Geohagen, S. Das, Toxicol. Sci., 98 (2007)
561.
[27] R.M. LoPachin, T. Gavin, D.R. Petersen, D.S. Barber, Chem. Res. Toxicol.,
22 (2009) 1499.
[28] P. Jaramillo, P. Perez, R. Contreras, W. Tiznado, P. Fuentealba, J. Phys. Chem.
A, 110 (2006) 8181.
[29] C.M. Spickett, A.R. Pitt, N. Morrice, W. Kolch, Biochim. Biophys. Acta,
1764 (2006) 1823.
[30] F. Fenaille, P. Mottier, R.J. Turesky, S. Ali, P.A. Guy, J. Chromatogr. A, 921
(2001) 237.
[31] Gaussian, Revision C.02, Gaussian, Inc., Wallingford CT (2004).
[32] A.K. Yocum, T. Oe, A.L. Yergey, I.A. Blair, J. Mass Spectrom., 40 (2005)
754.
[33] X. Zhu, V.E. Anderson, L.M. Sayre, Rapid Commun. Mass Spectrom., 23
(2009) 2113.
62
[34] A. Furuhata, T. Ishii, S. Kumazawa, T. Yamada, T. Nakayama, K. Uchida, J.
Biol. Chem., 278 (2003) 48658.
[35] R.M. LoPachin, B.C. Geohagen, T. Gavin, Toxicol. Sci., 107 (2009) 171.
[36] B. Paizs, S. Suhai, Mass Spectrom. Rev., 24 (2005) 508.
[37] M.R. Roe, H. Xie, S. Bandhakavi, T.J. Griffin, Anal. Chem., 79 (2007) 3747.
[38] N. Rauniyar, S.M. Stevens, K. Prokai-Tatrai, L. Prokai, Anal. Chem., 81
(2009) 782.
[39] A.L. Isom, S. Barnes, L. Wilson, M. Kirk, L. Coward, V. Darley-Usmar, J.
Am. Soc. Mass Spectrom., 15 (2004) 1136.
[40] M.S. Bolgar, S.J. Gaskell, Anal. Chem., 68 (1996) 2325.
[41] N. Rauniyar, L. Prokai, Proteomics, 9 (2009) 5188.
Full citation for Reference [31]:
Gaussian 03, Revision C.02, Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.;
Scuseria, G. E.; Robb, M. A.; Cheeseman, J. R.; Montgomery, Jr., J. A.; Vreven,
T.; Kudin, K. N.; Burant, J. C.; Millam, J. M.; Iyengar, S. S.; Tomasi, J.; Barone,
V.; Mennucci, B.; Cossi, M.; Scalmani, G.; Rega, N.; Petersson, G. A.; Nakatsuji,
H.; Hada, M.; Ehara, M.; Toyota, K.; Fukuda, R.; Hasegawa, J.; Ishida, M.;
Nakajima, T.; Honda, Y.; Kitao, O.; Nakai, H.; Klene, M.; Li, X.; Knox, J. E.;
Hratchian, H. P.; Cross, J. B.; Bakken, V.; Adamo, C.; Jaramillo, J.; Gomperts, R.;
Stratmann, R. E.; Yazyev, O.; Austin, A. J.; Cammi, R.; Pomelli, C.; Ochterski, J.
W.; Ayala, P. Y.; Morokuma, K.; Voth, G. A.; Salvador, P.; Dannenberg, J. J.;
Zakrzewski, V. G.; Dapprich, S.; Daniels, A. D.; Strain, M. C.; Farkas, O.; Malick,
D. K.; Rabuck, A. D.; Raghavachari, K.; Foresman, J. B.; Ortiz, J. V.; Cui, Q.;
Baboul, A. G.; Clifford, S.; Cioslowski, J.; Stefanov, B. B.; Liu, G.; Liashenko, A.;
Piskorz, P.; Komaromi, I.; Martin, R. L.; Fox, D. J.; Keith, T.; Al-Laham, M. A.;
Peng, C. Y.; Nanayakkara, A.; Challacombe, M.; Gill, P. M. W.; Johnson, B.;
Chen, W.; Wong, M. W.; Gonzalez, C.; and Pople, J. A.; Gaussian, Inc.,
Wallingford CT, 2004.
63
Scheme 3.1. Lipid peroxidation products generated from PUFA, including
α,β-unsaturated aldehydes: 4-hydroxy-2-nonenal (HNE), 4-oxo-2-nonenal (ONE),
4-hydroxyhexanal (HHE), hexenal, crotonaldehyde, and acrolein (ACR) as well as
the dialdehydes: malondialdehyde (MDA) and glyoxal. They are formed by a
variety of mechanisms. For example, HNE can be produced through the
β-cleavage of hydroperoxides from ω-6 polyunsaturated fatty acids.
64
Scheme 3.2. Michael adducts formed from Cys-containing peptides with ACR,
ONE and HNE (A), C-2 can also be the active site for ONE; Schiff bases formed
from Lys-containing peptides with MDA and ONE (B).
Scheme 3.3. The aspartic acid effect explained by the charge-remote peptide
fragmentation pathway [36].
65
Scheme 3.4. Proposed pathways from Cys-containing peptide conjugate for (A)
neutral loss of HNE moiety according to a reversal Michael adduction mechanism,
and (B) neutral loss of HS-HNE (C9H18O2S) moiety according to a β-elimination
mechanism.
66
Table 3.1. Calculated ELUMO and EHOMO for three lipid peroxidation products and
amino acids. N and C termini of the amino acids are protected. The numbers
behind the amino acids indicate the charge state.
Table 3.2. Calculated nucleophilicity indexes (ω-) for lipid peroxidation product
reactions with possible nucleophilic targets. N and C termini of the amino acids
are protected. The numbers behind the amino acids indicate the charge state.
67
Table 3.3. Peptide adduction type by lipid peroxidation products under reaction
conditions: reaction time 2 h, reaction temperature 37 ℃, pH 7.4. Cys, His and
Lys are the reactive residue for the peptides. MA = Michael Adduct, SB = Schiff
Base, NCO = No Conjugate Observed.
Table 3.4. Characteristic fragment ions for oxylipid-peptide conjugates observed
in MALDI-MS/MS spectra.
68
Figure 3.1. MALDI MS demonstrating adduction type of the
Ac-SVVDLTHR-amide peptide by HNE. P annotates the original peptide; asterisk
* annotates the oxylipid labeling.
Figure 3.2. MALDI MS/MS spectrum of HNE-peptide conjugate (Michael adduct)
Ac-SVVDLTCR-amide. Precursor, P* = 1089.6 Th, L, a Leu immonium ion.
69
Figure 3.3. MALDI MS/MS spectrum of HNE-peptide conjugate (Michael adduct)
Ac-SVVDLTHR-amide. Precursor, P* = 1123.7 Th; H, a His immonium ion.
Figure 3.4. MALDI MS/MS spectrum of ONE-peptide conjugate (Michael adduct)
Ac-SVVDLTKR-amide. Precursor, P* = 1112.7 Th; K, a Lys immonium ion with
neutral loss of NH3.
70
Figure 3.5. MALDI MS/MS spectrum of MDA-peptide conjugate (Schiff base)
Ac-SVVDLTKR-amide. Precursor, P* = 1012.6 Th; K, a Lys immonium ion with
neutral loss of NH3.
71
Supporting information
Figure 3.S-1. MALDI MS/MS spectrum of ONE-peptide conjugate (Michael
adduct) Ac-SVVDLTCR-amide. P annotates the original peptide; asterisk *
annotates the oxylipid labeling. Precursor, P* = 1087.5 Th.
Figure 3.S-2. MALDI MS/MS spectrum of ACR-peptide conjugate (Michael
adduct) Ac-SVVDLTCR-amide. Precursor, P* = 989.4 Th.
72
Figure 3.S-3. MALDI MS/MS spectrum of ONE-peptide conjugate (Michael
adduct) Ac-SVVDLTHR-amide. Precursor, P* = 1121.7 Th; H, a His immonium
ion.
Figure 3.S-4. MALDI MS/MS spectrum of ACR-peptide conjugate (Michael
adduct) Ac-SVVDLTHR-amide. Precursor, P* = 1023.4 Th; H, a His immonium
ion.
73
Figure 3.S-5. MALDI MS/MS spectrum of HNE-peptide conjugate (Michael
adduct) Ac-SVVDLTKR-amide. Precursor, P* = 1114.7 Th; K, a Lys immonium
ion with neutral loss of NH3.
Figure 3.S-6. MALDI MS/MS spectrum of ACR-peptide conjugate (Michael
adduct) Ac-SVVDLTKR-amide. Precursor, P* = 1014.5 Th; K, a Lys immonium
ion with neutral loss of NH3.
74
Figure 3.S-7. MALDI MS/MS spectrum of ONE-peptide conjugate (Schiff base)
Ac-SVVDLTKR-amide. Precursor, P* = 1094.7 Th; K, a Lys immonium ion.
75
CHAPTER 4.
Mass Spectrometry-based Quantification of Acrolein-modified
Cys-containing Peptides in an In Vivo Model of Oxidative Stress
Jianyong Wu and Claudia S. Maier*
Department of Chemistry, Oregon State University, Corvallis OR, 97330
* Corresponding author.
Email: Claudia.maier@oregonstate.edu
Fax: 541-737-2062
76
Abstract
The adduction of proteins and other biomolecules by electrophilic lipid
peroxidation products such as 4-hydroxy-2-nonenal (HNE), 4-oxo-2-nonenal
(ONE) or acrolein (ACR) is thought to be an initiating and/or propagating factor
in the pathophysiology of several diseases such as atherosclerosis, diabetes,
Alzheimer’s, Parkinson’s and other age-related disorders. Determining the extent
or relative amounts of this oxidative damage could provide valuable insights into
the molecular mechanisms of these disorders. Relative quantitation of
oxylipid-modified proteins in biological samples is a challenging problem because
of the complexity and extreme dynamic range that characterize these samples. In
this study, the reagents, N'-aminooxymethylcarbonylhydrazino-D-biotin and
iodoacetyl-PEO2-biotin, were used to enrich acrolein-modified Cys-containing
peptides and the corresponding unmodified ones from mitochondrial samples. The
ratios between them were determined by nanoLC-SRM analysis. Model
Cys-containing peptides labeled with ARP-acrolein and IPB were employed to
demonstrate this method. Seven acrolein-modified Cys-containing peptides from
five mitochondrial proteins were quantified. The ratios for those seven peptides
from the CCl4 treated rats are higher than the control ones indicating that the
ratios of acrolein-modified peptides to unmodified ones are potential markers of
oxidative stress in vivo.
77
Introduction
Mitochondria are cell's powerhouse through oxidative phosphorylation. 2% [1] of
oxygen escapes in the respiration chain from mitochondria and forms superoxide
radical, a reactive oxygen species (ROS) which can initiate radical-mediated
oxygenations. Free-radical injury is crucial to age-related disorders and etiology
and pathology of inflammatory diseases. ROS is counteracted by an intricate
antioxidant system and the degree of oxidative stress is determined by the extent
of the imbalance of ROS production and antioxidant defenses [2]. Proteins, DNA
and lipids can be modified because of oxidative stress in biological system [2-6].
Oxidative modification of proteins occur according to a variety of mechanisms [3,
5, 7-9]. Some of them lead to backbone peptide bond cleavage or to side-chain
modifications. The latter introduces protein carbonyls by secondary reaction of
histidine, cysteine and lysine reside with lipid peroxidation products, which are
produced by polyunsaturated fatty acids and their esters under oxidative stress
[10-13]. 4-hydroxy-2-nonenal (HNE), 4-oxo-2-nonenal (ONE) and acrolein (ACR)
are the most reactive lipid peroxidation products. The adduction of proteins and
other biomoleculars by these lipid peroxidation products is thought to be an
initiating or propagating factor in the pathophysiology of several diseases, such as
atherosclerosis, cancer, diabetes-related complications and neurodegenerative
diseases [4, 14-19].
Traditionally, protein carbonyls were detected by approaches using 2,
78
4-dinitrophenylhydrazine (DNPH) [20] or anti-dinitrophenyl primary antibody [21,
22]. Recently, mass spectrometry is applied to identify the modification sites and
types for protein carbonyls in biological samples by different research groups
[23-25]. Since the protein carbonyls are only a small proportion from the complex
biological samples, an enrichment method is usually required before MS analysis.
For deeper understanding of impact of those modifications to cells, quantitative
analysis of those modified peptides is necessary. Quantification of peptides or
proteins based on mass spectrometry mainly has two approaches [26]. One major
approach is based on isotope-labeled amino acids or peptides, which includes
reagents like ICAT or iTRAQ, and methods like SILAC or AQUA. The other
major approach is label-free quantification including two different strategies: (i)
measuring and comparing signal intensity of peptides from proteins; (ii) counting
and comparing the number of identified-peptide spectra from a particular protein.
Quantification of protein carbonyls in biological samples are more challenging
because of the additional enrichment step involved. A few methods with stable
isotope labeling were carried out to measure peptides carbonyls in vitro in
different groups [27-29]. However, none of them determined the ratios between
oxylipid-modified peptides and unmodified peptides in biological sample in vivo,
which indicates the extent of oxylipid modification in cell. Here we introduced a
method that tried to reach this goal. In our previous study [23],
N'-aminooxymethylcarbonylhydrazino-D-biotin as an aldehyde reactive probe
79
(ARP) was used to enrich peptide carbonyls with avidin affinity columns, then the
peptide carbonyls were identified by mass spectrometry. We develop this method
with an additional iodoacetyl-PEO2-Biotin (IPB) reagent to determine the ratios of
seven acrolein-modified peptides to their corresponding unmodified peptides in
mitochondrial samples in vivo. In this method, ARP was used to react with
acrolein-modified peptides to form ARP-ACR modified peptides and IPB was
applied to react with unmodified peptides to form IPB labeled peptides (Scheme
4.1). The prepared samples were passed through a strong cation exchange (SCX)
column and an avidin affinity column for further nano-LC MS analysis. Selected
reaction monitoring (SRM) was employed for quantitative analysis because of its
high specificity and sensitivity in complex biological samples.
Material and methods
Materials
Sequencing grade-modified trypsin was purchased from Promega Corp. (Madison,
WI). Acrolein (ACR) (≥99%) was obtained from Fluka (St. Louis, MO). Aldehyde
reactive probe (ARP, N-aminooxymethylcarbonylhydrazino D-biotin) was
purchased
from
Dojindo
Laboratories
(Kumamoto,
Japan).
UltraLink-immobilized monomeric avidin, iodoacetyl-IPB2-Biotin (IPB) and
triton X-100 detergent were obtained from Pierce (Rockford, IL). Bovine serum
albumin (BSA) was from Calbiochem (La, CA). Macrospin strong cation
80
exchange (SCX) columns were from Nest Group (Southboro, MA).
Synthesis of ARP-ACR and IPB Labeled Peptides
One ml of 2-mg BSA (20 mM phosphate buffer, pH 8.3) was reacted with 25 μl of
DTT (60 mM, 20 mM phosphate buffer, pH 8.3) at 95 °C for 10 min. One half of
the reduced BSA was then mixed with 200 μl IPB (20 mM, 50 mM Trice buffer,
pH 8.3). The mixture was kept in the dark at room temperature for 90 min to form
IPB labeled BSA. The additional IPB was removed by adding 30 μl DTT (60 mM,
20 mM phosphate buffer, pH 8.3). The other half of the reduced BSA was
reacted with 60 μl of ACR (80 mM, 20 mM phosphate buffer, pH 8.3) at room
temperature for 60 min. Almost all of the Cys residues would be adducted with
ACR and nearly non of the Lys or Arg residues would be adducted under this
reaction condition according to the high difference of reactivity between them
[30-33]. The excess ACR was removed by adding 40 μl DTT (60 mM, 20 mM
phosphate buffer, pH 8.3). This ACR labeled BSA was then reacted with 250 μl
ARP (30 mM, in H2O) at room temperature for 60 min. Both ARP-ACR and IPB
labeled BSA were digested by trypsin and passed through 10 KDa filter. The
peptide (YICDNQDTISSK) labeled with ARP-ACR and IPB, respectively, will be
used in the quantitative model study.
81
SCX Cleanup
Procedure was followed the manufacture’s instructions. Acetonitrile was used to
active the column and an elution buffer with 20 % acetonitrile (pH 3.0; 0.6 M KCl)
was applied to condition the column. A washing buffer with 20% acetonitrile (pH
3.0; 10mM KCl) was used to equilibrate the column. The mitochondrial peptide
sample (~pH 3.0, adjusted with H3PO4) was added into the column and washed
with washing buffer three times to remove all small molecules with biotin tag and
triton X-100 detergent, finally 350 µl elution buffer was added to release the
peptide sample.
Affinity Chromatography
200 µl of Ultralink monomeric avidin was packed into Handee Mini Spin
Columns which accommodate solvent addition with a Luer Lok syringe. The
column was washed with 1.5 ml of 10 mM NaH2PO4, pH 7.4.
Irreversible
binding sites, consisting of tetrameric avidin, were blocked by washing with 600
µl of 2 mM D-biotin.
To remove excess D-biotin the column was washed with 1
ml of 2 M Glycine-HCl pH2.8.
The column was then re-equilibrated by
washing with 2 ml of 2 x phosphate buffered saline (PBS, 20 mM NaH2PO4
300mM NaCl).
The mitochondrial peptide sample was then slowly added to the
affinity column. To remove non-labeled and non-specifically bound peptides the
column was washed with 1 ml 2 x PBS followed by 1 ml of 10 mM NaH2PO4 and
82
finally 1.5 ml of 50 mM NH4HCO3 20% CH3OH. The column was then rinsed
with 1 ml of MilliQ H 2O before eluting the ARP labeled peptides with 0.4%
formic acid 20% acetonitrile.
Collected fractions were then concentrated using a
freeze dryer and stored at -20oC before mass spectrometric analysis.
Nano-HPLC
An Ultimate LC Packing system (Dionex, Sunnyvale, CA) was used. Peptide
samples were loaded onto a 5 mm × 0.50 mm C18 trap cartridge at a washing
flow rate of 20 μL/min. After 4 min the trap cartridge was automatically switched
in-line with a 75 μm i.d. ×15 cm C18 PepMap 100 column (Dionex, Sunnyvale,
CA). Peptides were eluted with a gradient from 9% to 18% B over 90 min (using
solvent A: 1% acetonitrile with 0.1% formic acid; solvent B, 100% acetonitrile
with 0.1% formic acid) at a 0.260 μL/min.
Mass Spectrometry
All experiments were carried out on a 4000 Q-Trap hybrid tandem mass
spectrometer (AB/MDS SCIEX, Concord, Ontario, Canada) using nano-ESI
source. The electrospray voltage was set to 2300 V and the declustering potential
was 60 V. For identification of ARP labeled peptides, the precursor ion scan was
performed over a mass range of 400–1300 Th at 500 Th/s (Q1 and Q3 with unit
resolution). An enhanced product ion scan (MS/MS) would be performed to
83
identify it if an intensity of a precursor of 227 Th exceeds the threshold value of
1000 cps. The scan rate for MS/MS was set to 4000 Th/s. The SRM mode was run
with Q1 and Q3 set at unit resolution to increase the specificity. Each SRM
transition time is 30 ms. SRM collision energies were 50 eV and 51 eV for
ARP-ACR labeled and IPB labeled model peptides. All other SRM collision
energies for peptides from mitochondrial samples were listed in Table 4.2.
Animal Treatment
The experimental protocol for the animal studies was approved by the
Institutional Animal Care and Use Committee at Oregon State University. Seven
Male F344 rats (Harlan, Indianapolis, IN) were housed in individual plastic cages
covered with Hepa filter and allowed free access to standard animal chow and
water ad libitum. After 1 week of acclimatization, one 25-month old rat was killed
for reproducibility study and the other six 3-month old rats were transferred to
metabolism cages. These six animals were divided into two groups of three, with
one group receiving an intraperitoneal dose of 1 ml/kg CCl4 (dissolved in corn oil),
and the other group (control) receiving the vehicle alone. The CCl4 dose of 1
ml/kg was chosen on the basis of literature reports [34, 35]. The rats were killed
24 h after the treatment.
84
Preparation of ARP-ACR and IPB labeled Mitochondrial Peptides for SRM
analysis
Rat cardiac mitochondria were isolated by differential centrifugation and stored at
-80 oC [36]. Each sample of subsarcolemmal mitochondria (SSM) containing
approximately 0.5 mg total protein was washed twice with 10 mM NaH2PO4 pH
7.4 at 0 oC. The mitochondria were then resuspended in 400 µl of 10 mM
NaH2PO4 pH 7.4 with 1% of triton X-100 detergent and 3 mM DTT. DTT was
used to react with free acrolein to prevent artificial results during sample
preparation.
The mitochondria were sonicated in ice water for 5 min.
Mitochondrial proteins were separated by centrifugation at 14 K × g for 15 min at
4 oC. This sample was then filtered through an Amicon Microcon centrifugal filter
(10 KDa MWCO) to remove the oxylipid, lipid and other small molecules at 4 oC.
Mitochondrial proteins were then resuspended in 400 µl of 10mM NaH2PO4 pH
7.4 with 1 mM DTT. Further sample preparation was followed in Scheme2. by
four steps.
(1) APR was added to sample to reach a final concentration of 3 mM and the
proteins were digested with a 1:50 ratio of trypsin at 37 oC for 15 h.
(2) 15 µl DTT (30 mM, 20 mM phosphate buffer, pH 8.0) was added to the
tryptic peptides to reduce all still existing disulfide bridge.
This sample was
then filtered through an Amicon Microcon centrifugal filter (10 KDa MWCO)
to remove trypsin and other big molecules or particles.
85
(3) 40 µl of this sample was added with 120 µl IPB (5 mM, 50 mM Tris buffer,
pH 8.3). This mixture was kept in dark at room temperature for 90 min to
form IPB labeled peptides. 80 µl DTT (30 mM, 20 mM phosphate buffer, pH
8.0) was added to remove the additional IPB. This sample was diluted by
water to reach 1750 µl.
(4) 35 µl of the sample from step 3 was mixed with 800 µl of the sample from
step 2 to form a new sample for nanoLC-SRM analysis after SCX column and
affinity column.
Result and discussion
CID-based Fragmentation of ARP-ACR and IPB Labeled Peptides
Low-energy collision-induced dissociation (CID) was used to obtain tandem mass
spectra for ARP-ACR and IPB labeled peptides. Fig. 4.1 showes spectra of doubly
charged EFNGLGDCLTK peptides with different modification from ADP/ATP
translocase 1 protein. The asterisk * means the ARP-ACR labeled residue and
pound
#
means IPB labeled one. The ions in Fig. 4.1 are mainly y and b ions,
which are typical for ESI low-energy CID. The Cys-containing fragment ions
always with ARP-ACR (+ 369.2 Da) or IPB (+ 414.2 Da) moiety indicated that
those two modifications only slightly change the fragmentation pattern of peptides
and peptide bonds are still the main bonds to be broken. The relative intensity
among those y or b ions is very close between the ARP-ACR and IPB labeled
86
peptides. There are two characteristic ions: F1 (227.1 Th) and F2 (332.1 Th) from
ARP in the ARP-ACR labeled peptide and also two ones: Fa (270.1 Th) and Fb
(332.2 Th) from IPB in the IPB labeled peptide. The similarity of the
fragmentation pattern makes SRM a possibly accurate method to compare these
differently modified peptides.
Identification of Oxylipid Modified Peptides from Mitochondrial Samples
Sample preparation for identification of oxylipid modified peptides was the same
as the previous study [23]. The sample was labeled with ARP and then eluted
from an avidin affinity column. However, a significant amount of peptides
without ARP labeling still exited after enrichment because of an overwhelming
amount of native peptides in the original sample. Since a strong fragment ion of
270.1 Th from ARP exits in tandem mass spectra of APR labeled peptides, a
precursor ion scan was carried out to trigger the tandem mass spectra with 4000
Q-trap for identification. Seven acrolein-modifed Cys-containing peptides labeled
as a through g from five mitochondrial proteins were identified (Table 4.1). More
oxylipid modified peptides were found in our group with more sensitive
instruments (Appendix B). To obtain better quality of spectra, a serial of
independent tandem spectra were achieved with known mass of precursor ions of
ARP-ACR and IPB Labeled Peptides. These tandem spectra are shown in Fig. 4.1
and Fig. 4.S-1-14. The collision energies and retention times will be used for
87
further quantitative analysis.
Standard Curves for Model ARP-ACR and IPB Labeled Peptides
An enrichment method is necessary to detect acrolein-modified peptides by mass
spectrometry in biological samples since the concentration of them is extremely
low. APR reagent with biotin tag was used to react with acrolein-modified
peptides and these labeled peptides would be captured by an avidin affinity
column. In order to compare them to unmodified ones, IPB also with biotin tag
was used to react with unmodified peptides. To test the reaction yields for those
two reactions, a Cys-containing model peptide reacted with IPB to form an IPB
labeled peptide; this model peptide also reacted with acrolein to form an
ACR-modified peptide, which further reacted with ARP to form an ARP-ACR
modified peptide. The MS spectra of these two crude products do not show the
peaks of the native peptide or the ACR-modified peptide indicating almost 100%
yield for these two reactions (more details in Chapter 2). Thus, the amounts of
ARP-ACR labeled peptide and IPB labeled peptides will represent those of
acrolein-modified ones and unmodified ones respectively.
Before those two reagents were applied to biological samples, the ARP-ACR and
IPB labeled model peptides (YICDNQDTISSK) digested from BSA protein was
tested first. The other tryptic peptides were considered matrix. The tandem mass
spectra for these two modified peptides were available in 4.S-1-2. Three fragments
88
were selected for LC-SRM transition analysis. Fig. 4.2 showed the SRM
chromatograms for the mixture of 100 fmol ARP-ACR labeled peptide and 100
fmol IPB labeled peptide. The SRM intensity areas from ARP-ACR labeled
peptide were very similar with those from IPB labeled peptide, which indicates
that the ionization efficiency of electrospray and fragmentation pattern for those
two modified peptides are very similar. To prove this point, three SRM standard
curves (Fig. 4.3) were built. The 100 fmol IPB labeled peptide was considered the
reference and the amount of ARP-ACR labeled peptide started from 10.0 fmol to
1000 fmol. The R2 was higher than 0.997, which indicated good linearity. The
response factors (slopes) were 0.979, 1.102 and 1.100, which were close to 1. To
simplify the later calculation, we assumed the response factor equaled to 1 thus
the ratio of intensity area from SRM transition can directly represent the amount
ratio between ARP-ACR and IPB labeled peptides. Considering of the extremely
high yield for formations of ARP-ACR and IPB labeled peptides, the ratios of
intensity area from SRM transitions can also directly represent the ratio of
acrorein-modified peptides to unmodified peptides. For more accurate calculation,
the ratios were needed to be multiplied by a factor, which was close to 1.
Quantification of Acrolein-modified
Mitochondrial Samples
Cys-containing
Peptides
in
Scheme 4.2 showed the sample preparation procedure, through which there were
89
only 0.800 µl (40×(35/1750) = 0.8) of tryptic peptide used to react with IPB to
form IPB labeled peptides to represent unmodified Cys-containing peptides and
800 µl tryptic peptide to react with ARP to form ARP-ACR labeled peptides to
represent acrolein-modified peptides. Thus, the ratio of ARP-ACR modified
peptides to IPB labeled peptides obtained from nanoLC-SRM analysis in mixed
sample should be divided by 1000 to achieve the ratio in original mitochondrial
samples. In this way, we can circumvent the huge concentration difference
between acrolein-modified peptides and their corresponding unmodified ones to
avoid problems like overloading the nano-LC column or saturation of detector of
the mass spectrometer.
From 14 tandem mass spectra of ARP-ACR and IPB labeled peptides (shown in
Fig. 4.1 and 4.S-3-14), the three circled fragments for each spectrum were applied
to SRM analysis. One of the SRM with the least interference was selected for
quantification. Those SRM chromatograms for seven ARP-ACR modified
peptides and their corresponding IPB labeled ones are shown in Fig. 4.4. From the
graph, we can see the ARP-ACR modified peptides usually are eluted a little
earlier than the corresponding IPB labeled ones and the peak shapes of
acrolein-modifed peptides usually are broader. To compensate the mass difference
(45 Da) between ARP-ACR and IPB labeled peptides, the collision energy for the
IPB labeled peptide is one eV higher than the corresponding ARP-ACR modified
one. Detail information of nano-LC and MS properties for APR-ACR and IPB
90
labeled peptides are shown in Table 4.2. Because of the long sample preparation
and delicate nano-LC involved, a test for precision was necessary. Three samples
from one 25-month old rat went through sample preparation and nano-LC for
SRM analysis. The results are shown in Table 4.3. The CV for these three samples
was around 5%. The ratio of acrolein-modified peptides to unmodified ones
varied from 0.04% to 0.1%. There is still no available published data to directly
compare with our results. Judge et al. [37] prepared SSM sample almost the same
way and obtained 0.26 nmol protein carbonyl/mg protein using an enzyme
immunoassay. All peptide carbonyls in mitochondria were identified to be
acrolein-modified and Cys-containing in our experiments, which was reasonable
considering much higher concentration of ACR in cell than other reactive
oxylipids (ONE, HNE) and much higher reactivity for Cys than His and Lys
[30-33]. If we use acrolein-modified Cys-containing peptides to replace peptide
carbonyls and assume that the mass percentage of Cys in proteins is 1.5% [38]
and the ratio of Cys for modification varies from 0.04% to 0.1%, the calculated
protein carbonyl amount will vary from 0.06 to 0.15 nmol protein carbonyl/mg
protein, which is comparable with the results from Judge et al. (0.26 nmol protein
carbonyl/mg protein).
91
Quantification of Acrolein-modified Cys-containing Peptides under Oxidative
Stress
For further demonstration of this quantitative method, rats exposed to CCl4 as an
established oxidative stress in vivo model [39] were used in experiments. In Fig.
4.5, the seven mean ratios of acrolein-modified peptides to unmodified ones for
CCl4 treated rats are higher than those for control rats. The difference of the ratios
for b, d, f, and g peptides is statistically significant with a t-test (p < 0.05). The
lack of statistical difference for the other three peptides could be due to the
individual variation for animals. Those results indicate that those ratios not only
represent the extent of modification for the peptides from acrolein but also could
serve as markers of oxidative stress in vivo.
Conclusions
In our study, a method to measure ratios of acrolein-modified Cys-containing
peptides to unmodified ones in mitochondrial samples in vivo was developed.
Seven acrolein-modifed Cys-containing peptides from five mitochondrial proteins
were quantified by nano-LC SRM analysis. ARP and IPB reagents were
successfully used to label our target peptides for enrichment. The dilution step
during sample preparation circumvented the huge concentration difference
between acrolein-modified peptides and their corresponding unmodified ones.
ARP-ACR and IPB labeled peptides show similar elctrospray ionization
92
efficiency and fragmentation pattern in low-energy CID, which makes LC-SRM
an appropriate way to measure the ratios between them. The ratios for those seven
peptides from the CCl4 treated rats are higher than the control ones, which
indicates the ratios of acrolein-modified peptides to unmodified ones are potential
markers of oxidative stress in vivo. This quantification method can also be
possibly applied to protein carbonyls in bodily fluids and tissues in oxidative
stress-related diseases.
Acknowledgements
This work was supported by grants from the NIH/NIA (AG025372). The mass
spectrometry core facility of the Environmental Health Sciences Center at Oregon
State University is supported in part by a grant from NIEHS (ES00210).
93
References
[1] A. Boveris, N. Oshino, B. Chance, Biochem. J., 128 (1972) 617.
[2] T. Finkel, N.J. Holbrook, Nature, 408 (2000) 239.
[3] R.L. Levine, E.R. Stadtman, Exp. Gerontol., 36 (2001) 1495.
[4] L.J. Marnett, J.N. Riggins, J.D. West, J. Clin. Invest., 111 (2003) 583.
[5] E.R. Stadtman, Ann. N. Y. Acad. Sci., 928 (2001) 22.
[6] J.D. West, L.J. Marnett, Chem. Res. Toxicol., 19 (2006) 173.
[7] B.S. Berlett, E.R. Stadtman, J. Biol. Chem., 272 (1997) 20313.
[8] R.T. Dean, S. Fu, R. Stocker, M.J. Davies, Biochem. J., 324 ( Pt 1) (1997) 1.
[9] R.L. Levine, Free Radic. Biol. Med., 32 (2002) 790.
[10] H. Esterbauer, M. Dieber-Rotheneder, G. Waeg, G. Striegl, G. Jurgens, Chem.
Res. Toxicol., 3 (1990) 77.
[11] H. Esterbauer, R.J. Schaur, H. Zollner, Free Radic. Biol. Med., 11 (1991) 81.
[12] C. Schneider, K.A. Tallman, N.A. Porter, A.R. Brash, J. Biol. Chem., 276
(2001) 20831.
[13] L.M. Sayre, D. Lin, Q. Yuan, X. Zhu, X. Tang, Drug Metab. Rev., 38 (2006)
651.
[14] R. Bucala, A. Cerami, Adv. Pharmacol., 23 (1992) 1.
[15] J.M. Gutteridge, B. Halliwell, Trends Biochem. Sci., 15 (1990) 129.
[16]W. Palinski, M.E. Rosenfeld, S. Yla-Herttuala, G.C. Gurtner, S.S. Socher, S.W.
Butler, S. Parthasarathy, T.E. Carew, D. Steinberg, J.L. Witztum, Proc. Natl. Acad.
Sci. U. S. A., 86 (1989) 1372.
94
[17] N. Traverso, S. Menini, L. Cosso, P. Odetti, E. Albano, M.A. Pronzato, U.M.
Marinari, Diabetologia, 41 (1998) 265.
[18] K. Uchida, Free Radic. Biol. Med., 28 (2000) 1685.
[19] G. Jurgens, J. Lang, H. Esterbauer, Biochim. Biophys. Acta, 875 (1986) 103.
[20] R.L. Levine, D. Garland, C.N. Oliver, A. Amici, I. Climent, A.G. Lenz, B.W.
Ahn, S. Shaltiel, E.R. Stadtman, Methods Enzymol., 186 (1990) 464.
[21] M.A. Korolainen, G. Goldsteins, I. Alafuzoff, J. Koistinaho, T. Pirttila,
Electrophoresis, 23 (2002) 3428.
[22] T. Reinheckel, S. Korn, S. Mohring, W. Augustin, W. Halangk, L. Schild,
Arch. Biochem. Biophys., 376 (2000) 59.
[23] J. Chavez, J. Wu, B. Han, W.G. Chung, C.S. Maier, Anal. Chem., 78 (2006)
6847.
[24] M.R. Roe, H. Xie, S. Bandhakavi, T.J. Griffin, Anal. Chem., 79 (2007) 3747.
[25] S.M. Stevens, Jr., N. Rauniyar, L. Prokai, J. Mass Spectrom., 42 (2007) 1599.
[26] M. Bantscheff, M. Schirle, G. Sweetman, J. Rick, B. Kuster, Anal. Bioanal.
Chem., 389 (2007) 1017.
[27] B. Han, J.F. Stevens, C.S. Maier, Anal. Chem., 79 (2007) 3342.
[28] D.L. Meany, H. Xie, L.V. Thompson, E.A. Arriaga, T.J. Griffin, Proteomics, 7
(2007) 1150.
[29] H. Mirzaei, F. Regnier, J. Chromatogr. A, 1134 (2006) 122.
[30] J.A. Doorn, D.R. Petersen, Chem. Res. Toxicol., 15 (2002) 1445.
[31] J.A. Doorn, S.K. Srivastava, D.R. Petersen, Chem. Res. Toxicol., 16 (2003)
1418.
[32] D. Lin, H.G. Lee, Q. Liu, G. Perry, M.A. Smith, L.M. Sayre, Chem. Res.
Toxicol., 18 (2005) 1219.
95
[33] X. Zhu, L.M. Sayre, Chem. Res. Toxicol., 20 (2007) 165.
[34] H. Chung, D.P. Hong, J.Y. Jung, H.J. Kim, K.S. Jang, Y.Y. Sheen, J.I. Ahn,
Y.S. Lee, G. Kong, Toxicol. Appl. Pharmacol., 206 (2005) 27.
[35] R. Srinivasan, M.J. Chandrasekar,
Ethnopharmacol., 113 (2007) 284.
M.J.
Nanjan,
B.
Suresh,
J.
[36] J.W. Palmer, B. Tandler, C.L. Hoppel, J. Biol. Chem., 252 (1977) 8731.
[37] S. Judge, Y.M. Jang, A. Smith, T. Hagen, C. Leeuwenburgh, FASEB J., 19
(2005) 419.
[38] D. Voet, J.G. Voet, Biochemistry, John Wiley & Sons, Inc., Hoboken, 2004.
[39] B. Halliwell, Gutteridge, T., Free Radicals in Biology and Medicine, Oxford
University Press, London, 2007.
96
Scheme 4.1. Structures and formations of ARP-ACR modified and IPB-labeled
peptides.
Scheme 4.2. Relative quantitation procedure for acrolein-modified peptides in
mitochondrial protein samples.
97
Table 4.1. Summary of the seven acrolein-modified peptides labeled as a through
g from five mitochondrial proteins. Peptide sequences, biological and functional
assignments were made on the basis of information from the NCBI
(http://www.ncbi.nlm.nih.gov/Pubmed) and the UniPort Knowledgebase
(http://www.uniprot.org) websites.
Table 4.2. LC and MS properties for ARP-ACR (*) and IPB (#) labeled peptides.
98
Table 4.3. Reproducibility test for our method with three samples from one
25-month old rat.
99
Figure 4.1. Tandem mass spectra of the doubly charged Peptide a with different
modification from ADP/ATP translocase 1 protein by ESI Low-energy CID: A,
ARP-ACR modified (MH22+, 783.5 Th); B, IPB labeled (MH22+, 805.8 Th). The
circled fragments were used for SRM analysis.
100
Figure 4.2. SRM chromatograms for the ARP-ACR modified model peptide
(MH22+, 878.4 Th, 100 fmol) and IPB-labeled model peptide (MH22+, 900.9 Th,
100 fmol).
101
Figure 4.3. Calibration curves for the ARP-ACR modified model peptide with
three SRM transsitions; the IPB modified model peptide (100 fmol) was used as
the internal standard.
102
Figure 4.4. SRM chromatograms for ARP-ACR modified peptides and
PEO-labeled peptides from mitochondrial samples (* means ARP-ACR modified
and # means IPB labeled).
Figure 4.5. Ratios of acrolein-modified peptides to unmodified ones for control
rats (n = 3) and rats (n = 3) oxidatively stressed by CCl4. All results are presented
as the mean ± SEM. (Black bars indicate control rats and gray bars indicate CCl4
treated ones).
103
Supporting information
Figure 4.S-1. Tandem mass spectrum of the doubly charged model peptide with
ARP-ACR modification (MH22+, 878.4 Th). The circled fragments were used for
SRM analysis.
Figure 4.S-2. Tandem mass spectrum of the doubly charged model peptide with
IPB modification (MH22+, 900.9 Th). The circled fragments were used for SRM
analysis.
104
Figure 4.S-3. Tandem mass spectrum of doubly charged Peptide b with ARP-ACR
modification (MH22+, 978.9 Th). The circled fragments were used for SRM
analysis.
Figure 4.S-4. Tandem mass spectrum of doubly charged Peptide b with IPB
modification (MH22+, 1001.4 Th). The circled fragments were used for SRM
analysis.
105
Figure 4.S-5. Tandem mass spectrum of doubly charged Peptide c with ARP-ACR
modification (MH22+, 825.8 Th). The circled fragments were used for SRM
analysis.
Figure 4.S-6. Tandem mass spectrum of doubly charged Peptide c with IPB
modification (MH22+, 848.4 Th). The circled fragments were used for SRM
analysis.
106
Figure 4.S-7. Tandem mass spectrum of doubly charged Peptide d with ARP-ACR
modification (MH22+, 820.3 Th). The circled fragments were used for SRM
analysis.
Figure 4.S-8. Tandem mass spectrum of doubly charged Peptide d with IPB
modification (MH22+, 842.8 Th). The circled fragments were used for SRM
analysis.
107
Figure 4.S-9. Tandem mass spectrum of doubly charged Peptide e with ARP-ACR
modification (MH22+, 686.3 Th). The circled fragments were used for SRM
analysis.
Figure 4.S-10. Tandem mass spectrum of doubly charged Peptide e with IPB
modification (MH22+, 708.9 Th). The circled fragments were used for SRM
analysis.
108
Figure 4.S-11. Tandem mass spectrum of doubly charged Peptide f with
ARP-ACR modification (MH22+, 825.9 Th). The circled fragments were used for
SRM analysis.
Figure 4.S-12. Tandem mass spectrum of doubly charged Peptide f with IPB
modification (MH22+, 848.5 Th). The circled fragments were used for SRM
analysis.
109
Figure 4.S-13. Tandem mass spectrum of doubly charged Peptide g with
ARP-ACR modification (MH22+, 842.0 Th). The circled fragments were used for
SRM analysis.
Figure 4.S-14. Tandem mass spectrum of doubly charged Peptide g with IPB
modification (MH22+, 864.5 Th). The circled fragments were used for SRM
analysis.
110
CHAPTER 5.
Quantification of Peptide Adducts of Acrolein in Subsarcolemmal
Mitochondria in Aging Rats
Jianyong Wu, Cristobal Miranda, Kristina Jonsson and Claudia S. Maier
Department of Chemistry, Oregon State University, Corvallis OR, 97330
111
Abstract
The modification of proteins by lipid peroxidation products is thought to be an
underlying molecular factor in the pathogenesis of many degenerative diseases
and age related disorders. In myocardia, there are two distinct types of
mitochondria: interfibrillar mitochondria (IFM) and subsarcolemmal mitochondria
(SSM).
Here,
seven
acrolein-modifed
Cys-containing
peptides
in
five
mitochondrial proteins in SSM from young and old rats were quantified by
LC–SRM (selected reaction monitoring) analysis. Western blot analysis with
N’-aminooxymethylcarbonyl hydrazino-D-biotin as an aldehyde reactive probe
(ARP) and anti-ACR monoclonal antibody was also applied to measure the
protein carbonyls and FDP-lysine derivatives for SSM with age, respectively. The
tentative age-related increase of distinct target acrolein-modified peptides in SSM
was confirmed by the increase of FDP-lysine derivatives. However, total protein
carbonyls measurement using ARP in Western blot analysis did not conform to
this change suggesting that age-related changes in protein carbonyls are complex
and would benefit from more specific measurement protocols.
Introduction
Mitochondria are cell's powerhouse through oxidative phosphorylation. About 2%
[1] of oxygen escapes in the respiration chain from mitochondria and forms
superoxide radical, a reactive oxygen species (ROS) which can initiate
112
radical-mediated oxygenations. Proteins, DNA and lipids can be modified under
oxidative stress in biological system [2-6]. Oxidative modification of proteins
occurs according to a variety of mechanisms [3, 5, 7-9]. Some of them lead to
backbone peptide bond cleavage or to side-chain modifications. The latter
introduces protein carbonyls by Michael addition reactions between lipid
peroxidation products and histidine, cysteine or lysine reside. These oxylipids,
including acrolein (ACR), could be produced by polyunsaturated fatty acids and
their esters under oxidative stress [10-13].
In myocardia, there are two distinct types of mitochondria: interfibrillar
mitochondria (IFM) and subsarcolemmal mitochondria (SSM) [14]. The former
mitochondria are located between the myofibrils and the latter ones are clustered
beneath the plasma membrane. Judge et al. measured protein carbonyls in SSM
and IFM from rat heart using an enzyme immunoassay [15]. They found a
significant increase in oxidative stress levels in IFM with age, while little change
occurred for SSM. Fannin et al. also found that oxidative metabolism was
impaired with age in IFM and remained unaltered in SSM [16]. Recently, our
group developed a method to detect protein carbonyls in IFM by Western blot
analysis with N’-aminooxymethylcarbonyl hydrazino-D-biotin as an aldehyde
reactive probe (ARP) [17]. The ratios between acrolein-modified Cys-containing
peptides and the corresponding unmodified ones in SSM were also successfully
measured by using ARP and iodoacetyl-PEO2-Biotin (IPB) reagents with
113
nanoLC-SRM analysis (Chapter 4). Here we used this method to measure seven
acrolein-modifed Cys-containing peptides in five mitochondrial proteins. Western
blot with ARP and anti-ACR antibody was also applied to measure the protein
carbonyls and FDP-lysine derivatives, respectively, in SSM from young and old
rats.
Materials and methods
Materials
N’-aminooxymethylcarbonyl hydrazino-D-biotin (ARP) was purchased from
Dojindo Molecular Technologies Inc. (Rockville, MD). Ultralink Monomeric
Avidin and SuperSignal West Pico Chemiluminescent substrate were purchased
from
Thermo
Scientific
(Rockford,
IL).
Neutravidin-HRP,
UltraLink-immobilized monomeric avidin, iodoacetyl-IPB2-Biotin (IPB) and
Triton X-100 detergent were obtained from Pierce (Rockford, IL). Sequencing
grade modified trypsin was purchased from Promega (Madison, WI). Tris-HCl
(10%) Ready Gels were purchased from Bio-Rad.
Immobilon-P PVDF
membranes were purchased from Millipore. Goat anti-mouse IgG–HRP conjugate
was obtained from Bio-Rad Laboratories (Hercules, CA). Monoclonal
anti-acrolein IgG was from COSMO BIO Co. Ltd. (Carlsbad, CA)
114
Animal Treatment
The experimental protocol for the animal studies was approved by the
Institutional Animal Care and Use Committee at Oregon State University. Six
3-month and six 25-month male F344 rats (Harlan, Indianapolis, IN) were housed
in individual plastic cages covered with Hepa filter and allowed free access to
standard animal chow and water ad libitum. After 1 week of acclimatization, these
twelve rats were killed for study.
SDS-PAGE and Western Blot with Anti-ACR Antibody and ARP
Rat cardiac mitochondria were isolated by differential centrifugation and stored at
-80 oC [14]. Each sample of subsarcolemmal mitochondria (SSM) was washed
twice with 10 mM NaH2PO4, pH 7.4, at 0 oC.
The mitochondria were then
resuspended in 200 µl of 10 mM NaH2PO4, pH 7.4, with 1% of SDS detergent and
10 mM DTT. DTT was used to react with free acrolein to prevent artificial results
during sample preparation.
The mitochondria were sonicated at room
temperature for 5 min and then were mixed with Laemmli sample buffer (BioRad)
with 5% 2-mercaptoethanol. The proteins were denatured at 95 oC for 5 min. One
dimension SDS-PAGE was accomplished by loading about 15 µg of
mitochondrial protein per well of Bio-Rad 10%Tris-HCl ReadyMade gels and
running them at a constant 130V for 70 minutes. For visualization of the protein
bands the gels were incubated with Biosafe Coomassie dye for 1 hr.
115
Western blotting with anti-ACR antibody was accomplished by transferring the
proteins from the gel to a PVDF membrane by applying a constant current of 300
mA for 2 hours. The membrane was blocked in 5% non-fat milk in TBS-T (TBS
buffer with 0.1% Tween-20) and then reacted with monoclonal anti-acrolein IgG
for one hour. After extensive washing, the membrane was incubated with goat
anti-mouse IgG–HRP conjugate for 1 hr. After several additional wash steps, the
membrane was exposed to the West Pico Chemiluminescent substrate for 5
minutes and developed using Kodak BioMax X-Ray film.
For Western blotting with ARP, ARP was added to reach a final concentration of 5
mM to label the protein carbonyls before the proteins were totally denatured.
After one dimension SDS-PAGE, the proteins were transferred from the gel to a
PVDF membrane by applying a constant current of 300 mA for 2 hours. The blot
was then blocked in TBS buffer with 0.5% Tween-20 and then washed extensively
before being incubated with 40 ng/mL of HRP-NeutrAvidin for 1 hr. After several
additional wash steps, the membrane was exposed to the West Pico
Chemiluminescent substrate for 5 minutes and developed using Kodak BioMax
X-Ray film. For these two Western blottings, optical densities (OD) of all the
bands within a given lane from the stained gel and the membrane were calculated
with Kodak ID Image Analysis software.
116
Preparation of ARP-ACR and IPB labeled Mitochondrial Peptides for SRM
analysis
Because of the low abundance of ACR-modified mitochondrial proteins, an
enrichment method is necessary for SRM analysis. ARP and IPB were used to
react with ACR-modified peptides and Cys-containing peptides, respectively
(Scheme 5.1). The resulting products with biotin tags were able to be enriched by
an avidin column. The Scheme 5.2 shows the sample preparation procedure. More
details are in Chapter 4.
Nano-HPLC and Mass Spectrometry
An Ultimate LC Packing system and a 75 μm i.d. ×15 cm C18 PepMap 100
column (Dionex, Sunnyvale, CA) were used. Peptides were eluted with a gradient
from 9% to 18% B over 90 min (using solvent A: 1% acetonitrile with 0.1%
formic acid; solvent B, 100% acetonitrile with 0.1% formic acid) at a 0.260
μL/min. SRM analysis was carried out on a 4000 Q-Trap hybrid tandem mass
spectrometer (AB/MDS SCIEX, Concord, Ontario, Canada) using nano-ESI
source. More details are in Chapter 4.
Results and discussion
Quantification of Acrolein-modified Cys-containing Peptides in SSM for
Young and Old Rats by SRM analysis
Seven acrolein-modifed Cys-containing peptides from five mitochondrial proteins
117
were identified previously by mass spectrometry (Chapter 4). These modified
peptides can also represent acrolein-modifed Cys-containing peptides in SSM.
Table 5.1 shows these peptides labeled as a through g. Nano-LC SRM analysis
was used to quantify these seven modified peptides from six young (3-month) and
six old (25-month) rats. Fig. 5.1 shows the ratios of acrolein-modified peptides to
unmodified ones from young and old rats. The ratios were around 0.1%. There
was a tentative increase of ACR-modified peptides with age. However, statistical
analysis using a t-test (p-values < 0.05) indicating that there was no significant
difference between levels of ACR-modified peptides for young and old rats.
Measurement of Protein Carbonyls in SSM for Young and Old Rats by
Western Blot with ARP
As an alternative to the 2,4-dinitrophenylhydrazine-based assays for protein
carbonyls measurement, our group developed an approach, in which ARP reacted
with carbonyl group to form a biotinylated oxime derivatives [17]. After ARP
treatment, the SSM proteins were separated by 1D-SDS-PAGE followed by
Western blotting on a PVDF membrane. HRP conjugated NeutrAvidin was used
with PicoWest chemilumiinescent reagents to visualize the ARP reactive bands
(Fig. 5.2). Some ARP reactive protein bands are visible based on the Western blot.
The control sample without ARP addition shows relatively weak signals for two
bands. These two HRP-NeutrAvidin reactive protein bands in the control could be
118
due to naturally biotinylated proteins, such as one of the mitochondrial
carboxylase enzymes. Optical densities (OD) of all the bands within a given lane
from the stained gel and the Western blot membrane indicate the total amount of
protein and protein carbonyl, respectively. After correction of protein amount in
each lane, there is no significant difference of protein carbonyls in SSM based on
OD analysis with age in Fig. 5.3. Little concentration change of protein carbonyls
might be due to the undamaged anti-oxidative stress system in SSM with age [15].
Measurement of FDP-lysine Derivatives in SSM for Young and Old Rats by
Western Blot with Anti-ACR Antibody
Acrolein is α,β-unsaturated aldehyde, which can react with some side chains of
amino acids. One of the adducts is the formyl-dehydropiperidino-lysine
(FDP-lysine) derivative [18]. Anti-ACR monoclonal antibody is specific for
ACR-modified proteins, especially for FDP-lysine derivatives [19]. Here, we
applied this antibody to SSM samples. Fig. 5.4 shows the images of SDS-PAGE
and Western blot of FDP-lysine derivatives. After correction of protein amount
from SDS-PAGE image based on OD analysis, Fig. 5.5 implies the levels of
FDP-lysine derivatives in SSM in old rats are higher than young ones.
Conclusions
Ratios of ACR-modified Cys-containing peptides to unmodified ones in SSM with
119
age were successfully measured by LC-SRM analysis. A tentative increase of
distinct Michael adducts of ACR was observed, however statistical analysis
indicated that the observed difference in acrolein adduction was not significant (p
< 0.05) for the two age groups. Complementation of the LC-SRM analysis by
immunochemical analysis using an anti-ACR antibody showed that the levels of
FDP-lysine derivatives were higher in SSM isolated from old animals compared
to the levels determined for young animals. However, total protein carbonyls
measurement using ARP in Western blot analysis did not conform to this change
suggesting that age-related changes in protein carbonyls are complex and would
benefit from more specific measurement protocols.
Acknowledgements
This work was supported by grants from the NIH/NIA (AG025372). The mass
spectrometry core facility of the Environmental Health Sciences Center at Oregon
State University is supported in part by a grant from NIEHS (ES00210).
120
Reference
[1] A. Boveris, N. Oshino, B. Chance, Biochem. J., 128 (1972) 617.
[2] T. Finkel, N.J. Holbrook, Nature, 408 (2000) 239.
[3] R.L. Levine, E.R. Stadtman, Exp. Gerontol., 36 (2001) 1495.
[4] L.J. Marnett, J.N. Riggins, J.D. West, J. Clin. Invest., 111 (2003) 583.
[5] E.R. Stadtman, Ann. N. Y. Acad. Sci., 928 (2001) 22.
[6] J.D. West, L.J. Marnett, Chem. Res. Toxicol., 19 (2006) 173.
[7] B.S. Berlett, E.R. Stadtman, J. Biol. Chem., 272 (1997) 20313.
[8] R.T. Dean, S. Fu, R. Stocker, M.J. Davies, Biochem. J., 324 ( Pt 1) (1997) 1.
[9] R.L. Levine, Free Radic. Biol. Med., 32 (2002) 790.
[10] H. Esterbauer, M. Dieber-Rotheneder, G. Waeg, G. Striegl, G. Jurgens, Chem.
Res. Toxicol., 3 (1990) 77.
[11] H. Esterbauer, R.J. Schaur, H. Zollner, Free Radic. Biol. Med., 11 (1991) 81.
[12] C. Schneider, K.A. Tallman, N.A. Porter, A.R. Brash, J. Biol. Chem., 276
(2001) 20831.
[13] L.M. Sayre, D. Lin, Q. Yuan, X. Zhu, X. Tang, Drug Metab. Rev., 38 (2006)
651.
[14] J.W. Palmer, B. Tandler, C.L. Hoppel, J. Biol. Chem., 252 (1977) 8731.
[15] S. Judge, Y.M. Jang, A. Smith, T. Hagen, C. Leeuwenburgh, FASEB J., 19
(2005) 419.
[16] S.W. Fannin, E.J. Lesnefsky, T.J. Slabe, M.O. Hassan, C.L. Hoppel, Arch.
Biochem. Biophys., 372 (1999) 399.
[17] W.G. Chung, C.L. Miranda, C.S. Maier, Electrophoresis, 29 (2008) 1317.
121
[18] K. Uchida, M. Kanematsu, Y. Morimitsu, T. Osawa, N. Noguchi, E. Niki, J.
Biol. Chem., 273 (1998) 16058.
[19] K. Uchida, M. Kanematsu, K. Sakai, T. Matsuda, N. Hattori, Y. Mizuno, D.
Suzuki, T. Miyata, N. Noguchi, E. Niki, T. Osawa, Proc. Natl. Acad. Sci. U. S. A.,
95 (1998) 4882.
122
Scheme 5.1. Structures and formations of ARP-ACR modified and IPB-labeled
peptides.
Scheme 5.2. Relative quantitation procedure for acrolein-modified peptides in
mitochondrial protein samples.
123
Table 5.1. Summary of the seven acrolein-modifed peptides labeled as a through g
from five mitochondrial proteins. Peptide sequences, biological and functional
assignments were made on the basis of information from the NCBI
(http://www.ncbi.nlm.nih.gov/Pubmed) and the UniPort Knowledgebase
(http://www.uniprot.org) websites.
124
Figure 5.1. Ratios of acrolein-modified peptides to unmodified ones for young
rata (n = 6) and old rats (n = 6). All results are presented as the mean ± SEM.
(Green bars indicate young rats and blue bars indicate old ones).
125
Figure 5.2. Images of SDS-PAGE (A) and Western ARP blot of protein carbonyls
(B) for mitochondrial proteins from young (Y) and old (O) rats. (STD indicates
standard proteins and Con indicates control sample.)
126
Figure 5.3. Western ARP blot analysis of protein carbonyls in SSM from young
rats (n = 3) and old rats (n = 3). Optical densities (OD) of all the protein carbonyl
bands within a given lane were calculated with Kodak ID Image Analysis
software. All results are presented as the mean ± SEM.
127
Figure 5.4. Images of SDS-PAGE (A) and Western blot of FDP-lysine Derivative
(B) for mitochondrial proteins from young (Y) and old (O) rats. STD indicates
standard proteins.
128
Figure 5.5. Western blot analysis of FDP-lysine derivative in SSM from young
rats (n = 3) and old rats (n = 3). Optical densities (OD) of all the FDP-lysine
derivative bands within a given lane were calculated with Kodak ID Image
Analysis software. All results are presented as the mean ± SEM; p value is 0.05,
indicating the difference.
129
CHAPTER 6. CONCLUSIONS
MS/MS spectra under high-energy collision-induced dissociation (CID) allow
obtaining sequence information for oxylipid-peptide conjugates which, at least in
favorable cases, enables unambiguous localization of the modification sites by
lipid peroxidation products. The protonated dehydrated HNE moiety (139.1 Th) is
not shown in tandem mass spectra for His-containing peptide oxylipid conjugates,
compared with low energy CID. All MS/MS spectra of Michael adducts exhibit
neutral loss of the oxylipid moiety ions because they have –CH2CH2CH=O
structure in common. Spectra of Cys-containing peptide conjugates exhibit
additional characteristic neutral loss of HS-oxylipid moiety ions. In addition to
neutral loss of oxylipid moiety ions, spectra of His/Lys-containing peptide
conjugates (Michael adduct) show characteristic oxylipid-containing His/Lys
immonium or related ions. Spectra of Lys-containing peptide conjugates (Michael
adduct) show an additional Lys immonium ion with neutral loss of NH3. Spectra
of Lys-containing peptide conjugates (Schiff base) also show oxylipid-containing
Lys immonium ions or ions with neutral loss of NH3.
In our study, N'-aminooxymethylcarbonylhydrazino-D-biotin (ARP), as an
enrichment probe, was successfully used to identify seven acrolein-modifed
Cys-containing peptides from subsarcolemmal mitochondria (SSM) with a Q-trap
130
instrument. Combined with iodoacetyl-PEO2-Biotin (IPB) reagents, these seven
acrolein-modifed Cys-containing peptides were quantified by nano-LC SRM
analysis. ARP and IPB reagents were successfully used to label our target peptides
for enrichment. The dilution step during sample preparation circumvented the
huge concentration difference between acrolein-modified peptides and their
corresponding unmodified ones. ARP-ACR and IPB labeled peptides show similar
elctrospray ionization efficiency and fragmentation pattern in low-energy CID,
which makes LC-SRM an appropriate way to measure the ratios between them.
The ratios for those seven peptides from the CCl4 treated rats are higher than the
control ones, which indicates the ratios of acrolein-modified peptides to
unmodified
ones
are
potential
markers
of
oxidative
stress
in
vivo.
Age-dependent changes of protein carbonyls were also investigated in
subsarcolemmal mitochondria
by using this LC-SRM analysis of distinct
ACR-modified Cys-containing peptides. Immunochemical analysis using an
anti-ACR monoclonal antibody supported an increase of proteins modified by
acrolein with age. However, total protein carbonyls measurement using ARP in
Western blot analysis did not conform to this change suggesting that age-related
changes in protein carbonyls are complex and would benefit from more specific
measurement protocols.
131
ABBREVIATION
ACR
acrolein
ANT
adenine nucleotide transporter
AQUA
absolute quantification of protein
ARP
aldehyde reactive probe
B
magnetic sector
BSA
bovine serum albumin
CID
collision-induced dissociation
CV
coefficient of variation
DFT
density functional theory
DNPH
2, 4-dinitrophenylhydrazine
DTT
dithiothreitol
ECD
electron capture dissociation
ESI
electrospray ionization
ETC
electron transport chain
ETD
electron transfer dissociation
FMO
frontier molecular orbital
HNE
4-hydroxy-2-nonenal
HOMO
highest occupied molecular orbital
132
ABBREVIATION (Continued)
HRP
horseradish peroxidase
ICAT
isotope-coded affinity tag
ICR
ion cyclotron resonance
IFM
interfibrillar mitochondria
IPB
iodoacetyl-PEO2-Biotin
iTRAQ
isotope tags for relative and absolute quantification
LIT
linear ion trap
LUMO
lowest unoccupied molecular orbital
MALDI
matrix assisted laser desorption ionization
MCP
micro-channel plate
MDA
malondialdehyde
MPT
mitochondrial permeability transition
OD
optical density
ONE
4-oxo-2-nonenal
PBS
phosphate buffered saline
PTM
post translational modification
PUFA
polyunsaturated fatty acid
PVDF
polyvinylidene fluoride
Q
quadrupole
QIT
quadrupole ion trap
133
ABBREVIATION (Continued)
RF
radio frequency
ROS
reactive oxygen species
SCX
strong cation exchange
SEM
standard error of the mean
SILAC
stable isotope labeling by amino acids in cell culture
SOD
superoxide dismutase
SRM
selected reaction monitoring
SSM
subsarcolemmal mitochondria
TBARS
thiobarbituric acid
TDC
time-to-digital converter
TOF
time-of-flight
VDAC
voltage dependent anion selective channel
134
BIBLIOGRAPHY
[1] R. Aebersold, M. Mann, Nature, 422 (2003) 198.
[2] N.L. Anderson, N.G. Anderson, Electrophoresis, 19 (1998) 1853.
[3] A.Y. Andreyev, Y.E. Kushnareva, A.A. Starkov, Biochemistry (Mosc), 70
(2005) 200.
[4] M. Bantscheff, M. Schirle, G. Sweetman, J. Rick, B. Kuster, Anal. Bioanal.
Chem., 389 (2007) 1017.
[5] J.S. Beckman, W.H. Koppenol, Am. J. Physiol., 271 (1996) C1424.
[6] B.S. Berlett, E.R. Stadtman, J. Biol. Chem., 272 (1997) 20313.
[7] R.J. Beynon, M.K. Doherty, J.M. Pratt, S.J. Gaskell, Nat. Methods, 2 (2005)
587.
[8] W.P. Blackstock, M.P. Weir, Trends Biotechnol., 17 (1999) 121.
[9] M.S. Bolgar, S.J. Gaskell, Anal. Chem., 68 (1996) 2325.
[10] P.V. Bondarenko, D. Chelius, T.A. Shaler, Anal. Chem., 74 (2002) 4741.
[11] A. Boveris, B. Chance, Biochem. J., 134 (1973) 707.
[12] A. Boveris, N. Oshino, B. Chance, Biochem. J., 128 (1972) 617.
[13] R. Bucala, A. Cerami, Adv. Pharmacol., 23 (1992) 1.
[14] D. Bylund, R. Danielsson, G. Malmquist, K.E. Markides, J. Chromatogr. A,
961 (2002) 237.
[15] M. Carini, G. Aldini, R.M. Facino, Mass Spectrom. Rev., 23 (2004) 281.
[16] J. Chavez, J. Wu, B. Han, W.G. Chung, C.S. Maier, Anal. Chem., 78 (2006)
6847.
[17] K.B. Choksi, W.H. Boylston, J.P. Rabek, W.R. Widger, J. Papaconstantinou,
Biochim. Biophys. Acta, 1688 (2004) 95.
135
[18] H. Chung, D.P. Hong, J.Y. Jung, H.J. Kim, K.S. Jang, Y.Y. Sheen, J.I. Ahn,
Y.S. Lee, G. Kong, Toxicol. Appl. Pharmacol., 206 (2005) 27.
[19] W.G. Chung, C.L. Miranda, C.S. Maier, Brain Res., 1176 (2007) 133.
[20] W.G. Chung, C.L. Miranda, C.S. Maier, Electrophoresis, 29 (2008) 1317.
[21] W.G. Chung, C.L. Miranda, J.F. Stevens, C.S. Maier, Food Chem. Toxicol., 47
(2009) 827.
[22] B.A. Collings, W.R. Stott, F.A. Londry, J. Am. Soc. Mass Spectrom., 14
(2003) 622.
[23] J.J. Coon, J.E. Syka, J. Shabanowitz, D.F. Hunt, Biotechniques, 38 (2005)
519.
[24] M. Crompton, Biochem. J., 341 ( Pt 2) (1999) 233.
[25] B. D'Autreaux, M.B. Toledano, Nat. Rev. Mol. Cell Biol., 8 (2007) 813.
[26] R.T. Dean, S. Fu, R. Stocker, M.J. Davies, Biochem. J., 324 ( Pt 1) (1997) 1.
[27] M.K. Dennehy, K.A. Richards, G.R. Wernke, Y. Shyr, D.C. Liebler, Chem.
Res. Toxicol., 19 (2006) 20.
[28] J.A. Doorn, D.R. Petersen, Chem. Res. Toxicol., 15 (2002) 1445.
[29] J.A. Doorn, S.K. Srivastava, D.R. Petersen, Chem. Res. Toxicol., 16 (2003)
1418.
[30] J.M. Duan, M. Karmazyn, Mol. Cell Biochem., 90 (1989) 47.
[31] H. Esterbauer, Am. J. Clin. Nutr., 57 (1993) 779S.
[32] H. Esterbauer, M. Dieber-Rotheneder, G. Waeg, G. Striegl, G. Jurgens, Chem.
Res. Toxicol., 3 (1990) 77.
[33] H. Esterbauer, R.J. Schaur, H. Zollner, Free Radic. Biol. Med., 11 (1991) 81.
[34] S.W. Fannin, E.J. Lesnefsky, T.J. Slabe, M.O. Hassan, C.L. Hoppel, Arch.
Biochem. Biophys., 372 (1999) 399.
136
[35] F. Fenaille, P. Mottier, R.J. Turesky, S. Ali, P.A. Guy, J. Chromatogr. A, 921
(2001) 237.
[36] J.B. Fenn, M. Mann, C.K. Meng, S.F. Wong, C.M. Whitehouse, Science, 246
(1989) 64.
[37] L. Fialkow, Y. Wang, G.P. Downey, Free Radic. Biol. Med., 42 (2007) 153.
[38] R. Fields, H.B. Dixon, Biochem. J., 121 (1971) 587.
[39] T. Finkel, N.J. Holbrook, Nature, 408 (2000) 239.
[40] B. Friguet, FEBS Lett, 580 (2006) 2910.
[41] B. Friguet, L.I. Szweda, FEBS Lett, 405 (1997) 21.
[42] A. Furuhata, T. Ishii, S. Kumazawa, T. Yamada, T. Nakayama, K. Uchida, J.
Biol. Chem., 278 (2003) 48658.
[43] A. Furuhata, M. Nakamura, T. Osawa, K. Uchida, J. Biol. Chem., 277 (2002)
27919.
[44] Gaussian, Revision C.02, Gaussian, Inc., Wallingford CT (2004).
[45] M. Geiszt, J.B. Kopp, P. Varnai, T.L. Leto, Proc. Natl. Acad. Sci. U. S. A., 97
(2000) 8010.
[46] S.A. Gerber, J. Rush, O. Stemman, M.W. Kirschner, S.P. Gygi, Proc. Natl.
Acad. Sci. U. S. A., 100 (2003) 6940.
[47] A. Gilchrist, C.E. Au, J. Hiding, A.W. Bell, J. Fernandez-Rodriguez, S.
Lesimple, H. Nagaya, L. Roy, S.J. Gosline, M. Hallett, J. Paiement, R.E. Kearney,
T. Nilsson, J.J. Bergeron, Cell, 127 (2006) 1265.
[48] N.M. Green, Biochem. J., 89 (1963) 585.
[49] S. Grimm, D. Brdiczka, Apoptosis, 12 (2007) 841.
[50] P.A. Grimsrud, M.J. Picklo, Sr., T.J. Griffin, D.A. Bernlohr, Mol. Cell
Proteomics, 6 (2007) 624.
[51] T. Grune, K.J. Davies, Mol. Aspects. Med., 24 (2003) 195.
137
[52] J.M. Gutteridge, B. Halliwell, Trends Biochem. Sci., 15 (1990) 129.
[53] S.P. Gygi, B. Rist, S.A. Gerber, F. Turecek, M.H. Gelb, R. Aebersold, Nat.
Biotechnol., 17 (1999) 994.
[54] J.W. Hager, Rapid Commun. Mass Spectrom., 16 (2002) 512.
[55] B. Halliwell, Gutteridge, T., Free Radicals in Biology and Medicine, Oxford
University Press, London, 2007.
[56] B. Han, J.F. Stevens, C.S. Maier, Anal. Chem., 79 (2007) 3342.
[57] J.D. Hayes, L.I. McLellan, Free Radic. Res., 31 (1999) 273.
[58] A.J. Heck, J. Krijgsveld, Expert. Rev. Proteomics, 1 (2004) 317.
[59] Y. Huang, S. Guan, H.S. Kim, A.G. Marshall, Int. J. Mass Spectrom. Ion
Processes, 152 (1996) 121.
[60] A.L. Isom, S. Barnes, L. Wilson, M. Kirk, L. Coward, V. Darley-Usmar, J.
Am. Soc. Mass Spectrom., 15 (2004) 1136.
[61] P. Jaramillo, P. Perez, R. Contreras, W. Tiznado, P. Fuentealba, J. Phys. Chem.
A, 110 (2006) 8181.
[62] S. Judge, Y.M. Jang, A. Smith, T. Hagen, C. Leeuwenburgh, FASEB J., 19
(2005) 419.
[63] G. Jurgens, J. Lang, H. Esterbauer, Biochim. Biophys. Acta, 875 (1986) 103.
[64] M. Karas, F. Hillenkamp, Anal. Chem., 60 (1988) 2299.
[65] R.A. Kohanski, M.D. Lane, Methods Enzymol., 184 (1990) 194.
[66] M.A. Korolainen, G. Goldsteins, I. Alafuzoff, J. Koistinaho, T. Pirttila,
Electrophoresis, 23 (2002) 3428.
[67] G. Kroemer, J.C. Reed, Nat. Med., 6 (2000) 513.
[68] A.J. Lambert, M.D. Brand, Methods Mol. Biol., 554 (2009) 165.
138
[69] M.R. Larsen, M.B. Trelle, T.E. Thingholm, O.N. Jensen, Biotechniques, 40
(2006) 790.
[70] R.L. Levine, Free Radic. Biol. Med., 32 (2002) 790.
[71] R.L. Levine, D. Garland, C.N. Oliver, A. Amici, I. Climent, A.G. Lenz, B.W.
Ahn, S. Shaltiel, E.R. Stadtman, Methods Enzymol., 186 (1990) 464.
[72] R.L. Levine, E.R. Stadtman, Exp. Gerontol., 36 (2001) 1495.
[73] D. Lin, H.G. Lee, Q. Liu, G. Perry, M.A. Smith, L.M. Sayre, Chem. Res.
Toxicol., 18 (2005) 1219.
[74] R.M. LoPachin, D.S. Barber, T. Gavin, Toxicol. Sci., 104 (2008) 235.
[75] R.M. LoPachin, T. Gavin, B.C. Geohagen, S. Das, Toxicol. Sci., 98 (2007)
561.
[76] R.M. LoPachin, T. Gavin, D.R. Petersen, D.S. Barber, Chem. Res. Toxicol.,
22 (2009) 1499.
[77] R.M. Lopachin, B.C. Geohagen, T. Gavin, Toxicol. Sci., 107 (2009) 171.
[78] J. Malmstrom, H. Lee, R. Aebersold, Curr. Opin. Biotechnol., 18 (2007) 378.
[79] B.A. Mamyrin, Int. J. Mass Spectrom. Ion Processes, 131 (1994) 1.
[80] K. Marley, D.T. Mooney, G. Clark-Scannell, T.T. Tong, J. Watson, T.M.
Hagen, J.F. Stevens, C.S. Maier, J. Proteome Res., 4 (2005) 1403.
[81] L.J. Marnett, J.N. Riggins, J.D. West, J. Clin. Invest., 111 (2003) 583.
[82] D.L. Meany, H. Xie, L.V. Thompson, E.A. Arriaga, T.J. Griffin, Proteomics, 7
(2007) 1150.
[83] H. Mirzaei, F. Regnier, Anal. Chem., 77 (2005) 2386.
[84] H. Mirzaei, F. Regnier, J. Chromatogr. A, 1134 (2006) 122.
[85] A. Nakamura, S. Goto, J. Biochem., 119 (1996) 768.
139
[86] S. Nemoto, K. Takeda, Z.X. Yu, V.J. Ferrans, T. Finkel, Mol. Cell Biol., 20
(2000) 7311.
[87] Y. Oda, K. Huang, F.R. Cross, D. Cowburn, B.T. Chait, Proc. Natl. Acad. Sci.
U. S. A., 96 (1999) 6591.
[88] M. Oh-Ishi, T. Ueno, T. Maeda, Free Radic. Biol. Med., 34 (2003) 11.
[89] H. Ohkawa, N. Ohishi, K. Yagi, Anal. Biochem., 95 (1979) 351.
[90] S.E. Ong, B. Blagoev, I. Kratchmarova, D.B. Kristensen, H. Steen, A. Pandey,
M. Mann, Mol. Cell Proteomics, 1 (2002) 376.
[91] S.E. Ong, M. Mann, Nat. Chem. Biol., 1 (2005) 252.
[92] S. Orrenius, V. Gogvadze, B. Zhivotovsky, Annu. Rev. Pharmacol. Toxicol.,
47 (2007) 143.
[93] R.E. Pacifici, K.J. Davies, Gerontology, 37 (1991) 166.
[94] B. Paizs, S. Suhai, Mass Spectrom. Rev., 24 (2005) 508.
[95]W. Palinski, M.E. Rosenfeld, S. Yla-Herttuala, G.C. Gurtner, S.S. Socher, S.W.
Butler, S. Parthasarathy, T.E. Carew, D. Steinberg, J.L. Witztum, Proc. Natl. Acad.
Sci. U. S. A., 86 (1989) 1372.
[96] J.W. Palmer, B. Tandler, C.L. Hoppel, J. Biol. Chem., 252 (1977) 8731.
[97] W. Paul, Angew. Chem., 102 (1990) 780.
[98] D.N. Perkins, D.J. Pappin, D.M. Creasy, J.S. Cottrell, Electrophoresis, 20
(1999) 3551.
[99] A. Pompella, A. Visvikis, A. Paolicchi, V. De Tata, A.F. Casini, Biochem.
Pharmacol., 66 (2003) 1499.
[100]
A. Rasola, P. Bernardi, Apoptosis, 12 (2007) 815.
[101]
N. Rauniyar, L. Prokai, Proteomics, 9 (2009) 5188.
[102] N. Rauniyar, S.M. Stevens, Jr., L. Prokai, Anal. Bioanal. Chem., 389
(2007) 1421.
140
[103] N. Rauniyar, S.M. Stevens, K. Prokai-Tatrai, L. Prokai, Anal. Chem., 81
(2009) 782.
[104] T. Reinheckel, S. Korn, S. Mohring, W. Augustin, W. Halangk, L. Schild,
Arch. Biochem. Biophys., 376 (2000) 59.
[105] J.R. Requena, M.X. Fu, M.U. Ahmed, A.J. Jenkins, T.J. Lyons, J.W.
Baynes, S.R. Thorpe, Biochem. J., 322 ( Pt 1) (1997) 317.
[106]
3747.
M.R. Roe, H. Xie, S. Bandhakavi, T.J. Griffin, Anal. Chem., 79 (2007)
[107] P.L. Ross, Y.N. Huang, J.N. Marchese, B. Williamson, K. Parker, S.
Hattan, N. Khainovski, S. Pillai, S. Dey, S. Daniels, S. Purkayastha, P. Juhasz, S.
Martin, M. Bartlet-Jones, F. He, A. Jacobson, D.J. Pappin, Mol. Cell Proteomics,
3 (2004) 1154.
[108] L.M. Sayre, D. Lin, Q. Yuan, X. Zhu, X. Tang, Drug Metab. Rev., 38
(2006) 651.
[109]
L.M. Sayre, M.A. Smith, G. Perry, Curr. Med. Chem., 8 (2001) 721.
[110] C. Schneider, K.A. Tallman, N.A. Porter, A.R. Brash, J. Biol. Chem., 276
(2001) 20831.
[111] J.C. Schwartz, M.W. Senko, J.E. Syka, J. Am. Soc. Mass Spectrom., 13
(2002) 659.
[112] B.A. Soreghan, F. Yang, S.N. Thomas, J. Hsu, A.J. Yang, Pharm. Res., 20
(2003) 1713.
[113] C.M. Spickett, A.R. Pitt, N. Morrice, W. Kolch, Biochim. Biophys. Acta,
1764 (2006) 1823.
[114] R. Srinivasan, M.J. Chandrasekar, M.J. Nanjan, B. Suresh, J.
Ethnopharmacol., 113 (2007) 284.
[115]
E.R. Stadtman, Ann. N. Y. Acad. Sci., 928 (2001) 22.
[116]
W.E. Stephens, Phys. Rev., 69 (1946) 691.
[117]
W.E. Stephens, Bull. Am. Phys. Soc., 21 (1946) 22.
141
[118]
1599.
S.M. Stevens, Jr., N. Rauniyar, L. Prokai, J. Mass Spectrom., 42 (2007)
[119] J. St-Pierre, J.A. Buckingham, S.J. Roebuck, M.D. Brand, J. Biol. Chem.,
277 (2002) 44784.
[120] N. Tanaka, S. Tajima, A. Ishibashi, K. Uchida, T. Shigematsu, Arch.
Dermatol. Res., 293 (2001) 363.
[121]
1172.
A. Temple, T.Y. Yen, S. Gronert, J. Am. Soc. Mass Spectrom., 17 (2006)
[122] D. Toroser, W.C. Orr, R.S. Sohal, Biochem. Biophys. Res. Commun., 363
(2007) 418.
[123] N. Traverso, S. Menini, L. Cosso, P. Odetti, E. Albano, M.A. Pronzato,
U.M. Marinari, Diabetologia, 41 (1998) 265.
[124]
Y. Tsujimoto, S. Shimizu, Apoptosis, 12 (2007) 835.
[125]
K. Uchida, Free Radic. Biol. Med., 28 (2000) 1685.
[126] K. Uchida, M. Kanematsu, Y. Morimitsu, T. Osawa, N. Noguchi, E. Niki,
J. Biol. Chem., 273 (1998) 16058.
[127] K. Uchida, M. Kanematsu, K. Sakai, T. Matsuda, N. Hattori, Y. Mizuno,
D. Suzuki, T. Miyata, N. Noguchi, E. Niki, T. Osawa, Proc. Natl. Acad. Sci. U. S.
A., 95 (1998) 4882.
[128]
2004.
D. Voet, J.G. Voet, Biochemistry, John Wiley & Sons, Inc., Hoboken,
[129]
242.
M.P. Washburn, D. Wolters, J.R. Yates, 3rd, Nat. Biotechnol., 19 (2001)
[130]
J.D. West, L.J. Marnett, Chem. Res. Toxicol., 19 (2006) 173.
[131]
W.C. Wiley, I.H. McLaren, Rev. Sci. Instrum., 26 (1955) 1150.
[132]
H.L. Wong, D.C. Liebler, Chem. Res. Toxicol., 21 (2008) 796.
[133]
J.C. Yang, G.A. Cortopassi, Free Radic. Biol. Med., 24 (1998) 624.
142
[134] J.R. Yates, C.I. Ruse, A. Nakorchevsky, Annu. Rev. Biomed. Eng., 11
(2009) 49.
[135]
754.
A.K. Yocum, T. Oe, A.L. Yergey, I.A. Blair, J. Mass Spectrom., 40 (2005)
[136]
B.S. Yoo, F.E. Regnier, Electrophoresis, 25 (2004) 1334.
[137] X. Zhu, V.E. Anderson, L.M. Sayre, Rapid Commun. Mass Spectrom., 23
(2009) 2113.
[138]
X. Zhu, L.M. Sayre, Chem. Res. Toxicol., 20 (2007) 165.
[139]
R.A. Zubarev, Curr. Opin. Biotechnol., 15 (2004) 12.
143
Appendix A
A New Role for an Old Probe: Affinity Labeling of Oxylipid Protein
Conjugates by N’-Aminooxymethylcarbonylhydrazino D-biotin
Juan Chavez, Jianyong Wu, Bingnan Han, Woon-Gye Chung and Claudia S.
Maier*
Department of Chemistry, Oregon State University, Corvallis, OR 97331
Analytical Chemistry
American Chemical Society
1155 Sixteenth Street N.W., Washington, DC, 20036
J. Chavez, J. Wu, B. Han, W.G. Chung, C.S. Maier, Anal Chem, 78 (2006) 6847
144
Abstract
Free radicals, electrophiles and endogenous reactive intermediates are generated
during normal physiological processes, and are capable of modifying DNA, lipids
and proteins. However, elevated levels of oxidative modifications of proteins by
reactive species are implicated in the etiology and pathology of oxidative
stress-mediated
diseases,
neurodegeneration
and
aging.
A
mass
spectrometry-based approach is reported that aids to the identification and
characterization
of
carbonyl
modified
proteins.
The
method
uses
N'-aminooxymethylcarbonylhydrazino-D-biotin, a biotinylated hydroxylamine
derivative, that forms an oxime derivative with the aldehyde/keto group found in
oxidatively modified proteins. In this paper, the method is demonstrated for one
class of carbonyl-modified proteins, namely oxylipid peptide and protein
conjugates
formed
by
Michael
addition-type
conjugation
reactions
of
α,β-unsaturated aldehydic lipid peroxidation products with nucleophilic peptide
side chains. This new application of an “old” probe, which has been used for the
detection of abasic sites in DNA strands, introduces a biotin moiety into the
oxylipid peptide conjugate. The biotin-modified oxylipid peptide conjugates is
then amenable to enrichment using avidin affinity capture. The described method
represents an attractive alternative to hydrazine-based derivatization methods for
oxidized peptides and proteins because the reduction step necessary for the
transformation of the hydrazone bond to the chemically more stable hydrazine
145
bond can be omitted. Tandem mass spectrometry of the labeled oxylipid peptide
conjugates indicates that the biotin moiety is at least partially retained on the
fragment ion during the collisionally induced dissociation experiments, a
prerequisite for the use of automated database searching of uninterpreted tandem
mass spectra. The reported approach is outlined for the detection, identification
and characterization of oxylipid peptide conjugates but the labeling chemistry
may also be applicable to other carbonyl-modified proteins.
Introduction
Free-radical injury is central to the etiology and pathology of inflammatory
diseases and age-related disorders. The burden of reactive oxygen species (ROS)
is counteracted by an intricate antioxidant system, which includes antioxidant
enzymes, e.g. SOD, catalase and glutathione peroxidase, and non-enzymatic, low
molecular weight antioxidants, perhaps most importantly glutathione and ascorbic
acid. However, the extent of the imbalance of ROS production and antioxidant
defenses determines the degree of oxidative stress [1]. Oxidative-stress mediated
modifications occur on DNA, proteins, and lipids [1-5]. Oxidative modification of
proteins proceeds according to variety of mechanisms, [2, 5-7] some of which
lead to backbone peptide bond cleavage or to side chain modifications. The latter
includes the introduction of carbonyl functionalities, i.e. aldehyde or keto groups,
by direct oxidation of most residues or via conjugation reactions of reactive
146
secondary oxidation products that arise, for example, from lipid peroxidation
processes [7].
Polyunsaturated fatty acids of membrane lipids are highly susceptible to
peroxidation by oxygen radicals. Exposure of linoleic acid, the major ω-6
polyunsaturated fatty acid in mammalian tissues, to reactive oxygen species leads
to the formation of lipid peroxidation (LPO) products. In Esterbauer’s studies on
the production of cytotoxic compounds in LPO processes, 4-hydroxy-2-nonenal
(HNE) was described as a major LPO product [8, 9]. To date, HNE is the best
studied LPO product with cytotoxic potential. LPO products that contain
α,β-unsaturated aldehyde/keto functionalities are highly reactive electrophiles that
react readily with nucleophilic sites in proteins, such as cysteine sulfhydryls,
histidine imidazole moieties and with ε-amino groups of lysine residues, forming
Michael addition-type conjugates and Schiff’s base-derived products [4,10].
Oxylipid-protein conjugates have been implicated in the pathogenesis of
inflammatory, [11] neurodegenerative and age-related diseases [12, 13].
Consequently, there is a high interest in analytical methodology that are capable of
identifying and characterizing oxidative protein modifications.
Traditionally, the total amount of protein-bound carbonyls was determined
colorimetrically [14]. Many protocols for detecting oxidative protein modifcations
are based on 2,4-dinitrophenylhydrazine (DNPH) derivatization followed by UV
spectroscopy [15] or immunochemical detection using an anti-dinitrophenyl
147
primary antibody [16,17]. Although these methods are capable of detecting
oxidatively modified proteins with high sensitivity, they fail to provide
information such as the identity of the modified proteins, the chemical nature and
site of the modification. Mass spectrometry-based approaches are, at least in part,
capable of overcoming some of these latter limitations. For instance,
matrix-assisted laser desorption/ionization time-of-flight (MALDI-ToF) mass
spectrometry was used for identifying HNE-modified lysine residues in tryptic
digests of in vitro modified glucose-6-phosphate dehydrogenase [18]. Also, the
characterization of HNE-modified peptides directly from unfractionated digests of
model proteins after using DNPH as MALDI matrix has been reported [19]. In
1996, Bolgar and Gaskell reported the detection of HNE-modified His-residues of
oxidized low density lipoprotein by electrospray ionization (ESI) tandem mass
spectrometry (MS/MS) [20]. Since then, numerous in vitro studies have been
reported using ESI-MS/MS for identifying the site of modifications by LPO
products in proteins. For example, ESI-MS/MS was employed for detecting HNE
modifications in hemoglobin, [21] subunit VIII of bovine heart cytochrome c
oxidase, [22] glyceraldehyde-3-phosphate dehydrogenase, [23] the heat shock
proteins hsp72 and hsp90, [24, 25] and protein disulfide isomerase [26]. Labeling
of carbonyl groups formed by metal-catalyzed oxidation of amino acid side chains
by biotin hydrazine for both gel-based and mass spectrometric detection have
been reported recently [27-29].
148
As a new alternative to hydrazine-based reagents for MS-based characterization
strategies of oxidative stress-induced protein modifications, this study reports the
use
of
N'-aminooxymethylcarbonylhydrazino
D-biotin,
a
biotinylated
hydroxylamine derivative which reacts with aldehyde/keto groups found in
oxidatively modified proteins under formation of an oxime derivative which is
amenable to tandem mass spectral analysis [30]. This aldehyde-reactive probe
(ARP) is commonly used for the detection of DNA modifications that resulted in
the formation of an aldehyde group from the oxidative damage on the abasic (AP)
sites of DNA strands [31, 32]. The hydroxylamine moiety of ARP forms a stable
C=N bond with the aldehyde functionality of the Michael-type oxylipid conjugate
(Scheme 1). Thus, the reduction step which is necessary for the stabilization of the
hydrazone bond formed during the derivatization of the protein carbonyl with
hydrazide-based reagents becomes obsolete. The derivatization of oxidized
proteins by ARP enables the use of avidin affinity selection protocols for
enrichment of ARP-labeled peptides with oxidative modification. To demonstrate
the applicability of the ARP approach for the characterization of oxidatively
modified proteins we focus in this paper on one group of carbonylated proteins,
namely Michael addition-type oxylipid conjugates.
Specifically, we describe the derivatization of HNE-modified peptide conjugates
by ARP, and the selective affinity enrichment and MS-based detection of
ARP-labeled peptides from tryptic digests of HNE-conjugated thioredoxin (TRX).
149
ARP-labeled oxylipid peptide conjugates were studied by tandem mass
spectrometry. Successful database searching of uninterpreted mass spectra of
ARP-labeled HNE-conjugated peptides was also performed demonstrating the
potential of ARP as a new adjunct to the analysis of protein carbonyls.
Exeperimental section
Materials
Thioredoxin and sequencing grade-modified trypsin were from Promega
Corporation (Madison, WI). HNE (10 mg/mL in ethanol) was obtained from
Cayman Chemical (Ann Arbor, MI).
Aldehyde Reactive Probe (ARP,
N-aminooxymethylcarbonylhydrazino D-biotin) was purchased from Dojindo
Laboratories (Kumamoto, Japan).
UltraLink immobilized monomeric avidin
was obtained from Pierce (Rockford, IL).
Methods.
Synthesis, modification with HNE and labeling with ARP of the peptide
Ac-SVVDLTCR-amide.
The peptide Ac-Ser-Val-Val-Asp-Leu-Thr-Cys-Arg-amide(Ac-SVVDLTCR-amide)
was synthesized using a PS3 automated solid phase peptide synthesizer (Protein
Technologies,
Inc.,
Tucson,
AZ).
9-Fluoroenylmethoxy-carbonyl
(Fmoc)
derivatives of the amino acids were from SynPep Co. (Dublin, CA). The first
150
residue, i.e. the C-terminal residue of the final peptide product bound to the
Fmoc-Arg(pbf)-Rink Amide-MBHA resin was purchased from AnaSpec Inc. (San
Jose, CA).
One mg of the synthetic peptide in 1mL 10 mM phosphate buffer, pH 7.4 was
reacted with 13 µL HNE (10 mg/mL) at room temperature for two hours. The
HNE-adducted peptide was purified by HPLC on a C18 column (Vydac, 4.6 x 250
mm, 5 µm), lyophilized, and its mass confirmed by MALDI-MS. HNE-adducted
peptide (100 g) was reacted with 200 µL 10 mM ARP in 10 mM sodium
phosphate buffer, pH 7.4, for one hour at room temperature. The ARP-labeled
HNE-peptide conjugate was isolated by HPLC as described above and
lyophilized.
HNE modification and ARP-Labeling of E. coli Thioredoxin.
Adduction of thioredoxin (TRX) with HNE was accomplished by dissolving 1 mg
(82 nMole) TRX in 1 mL 10 mM sodium phosphate, pH 7.4, to which a 10-fold
molar excess (13 µL) of HNE (10 mg/ml) was added. The reaction mixture was
then incubated at 37 oC for 2 hr. The TRX-HNE adduct was purified from
unreacted HNE by reverse-phase HPLC using a C4 column (Vydac, 10 x 250 mm,
5 µm). The isolated product was lyophilized and analyzed by MALDI-MS.
Affinity labeling of TRX-HNE was carried out by the addition 250 µL of 10 mM
ARP to the TRX-HNE reconstituted in 750 µL 10 mM sodium phosphate, pH 7.4.
151
The reaction mixture was incubated at 37 oC for 3.5 hr. Unreacted ARP was
removed using the same HPLC method as described above. After lyophilization,
ARP-labeled HNE-modified TRX adduct was digested using trypsin at a ratio of
1:50 at 37 oC for 18 hr.
Affinity chromatography.
Ultralink monomeric avidin (100 µl) was packed into the tip of a glass pasteur
pipette plugged with glass wool.
The column was equilibrated to room
temperature and washed with five column volumes of PBS buffer.
The
irreversible binding sites were blocked by washing the column with three column
volumes of 2 mM D-biotin. Excess biotin was removed from the reversible
binding sites by washing the column with five column volumes of 0.1 M glycine,
pH 2.8.
The column was re-equilibrated by rinsing with five columns of PBS
buffer. One hundred microliters of TRX-HNE-ARP tryptic digest was loaded onto
the column and incubated for one hour. Non-labeled peptides were eluted from
the column by rinsing with 10 column volumes of PBS buffer and collected in
five 200 µL fractions.
The HNE-ARP labeled peptides were eluted using 10
column volumes of 0.3 % formic acid and were also collected in five 200 µL
fractions.
The fractions were analyzed by MALDI-MS/MS and ESI-MS/MS.
152
Oxidation of a Trp-containing peptide and ARP treatment
The model peptide QAKWRLQTL (from AnaSpec Inc., San Jose, CA) was
treated with H2O2 overnight and reaction was stopped by adding catalase. ARP
(20 molar excess, 10 mM sodium phosphate buffer, pH 7.4) was added to the
oxidized peptide. The reaction mixture was incubated at 37 oC for 1 hr.
Identification of carbonylated proteins in mitochondria.
Rat cardiac mitochondria were isolated by differential centrifugation and stored
at -80oC [33]. The mitochondria were ruptured by several freeze thaw cycles.
Protein concentration was determined by the Pierce-Coomassie protein assay.
Membrane and soluble proteins were reacted with 5mM ARP for 2 hr at 37oC.
Excess ARP was removed by buffer exchange using Amicon Microcon centrifugal
filters (10kDa MWCO). Proteins were digested with a 1:50 ratio of trypsin at 37
o
C overnight. The resulting tryptic peptides were filtered through Amicon
Microcon centrifugal filters (10kDa MWCO) to remove trypsin and membrane
fragments. The peptide solutions were passed through a monomeric avidin
cartridge (Applied Biosystems) following the manufacturer’s instructions to
enrich ARP-labeled peptides. The fraction containing the ARP-labeled peptides
was analyzed by nano-LC MALDI-TOF/TOF mass spectrometry. The MS/MS
spectra were interpreted both manually and in an automated fashion with Mascot
software.
153
Mass Spectrometry.
MALDI-MS/MS analysis was performed on an ABI 4700 Proteomics Analyzer
with TOF/TOF optics equipped with a 200-Hz frequency-tripled Nd:YAG laser
operating at a wavelength of 355 nm (Applied Biosystems, Inc., Framingham,
MA). Protein mass spectra were acquired in the linear mode, whereas all peptide
mass spectra were obtained in the reflector mode. For both operational modes the
accelerating voltage was set to 20 kV. For MS/MS experiments, the collision
energy, which is defined by the potential difference between the source
acceleration voltage (8 kV) and the floating collision cell (7 kV), was set at 1 kV.
The precursor ion was selected by operating the timed gate window with 3-10 Da.
Gas pressure (air) in the collision cell was set to 6 x 10-7 Torr. Fragment ions were
accelerated by 15 kV into the reflector. Data were acquired in a mass range of m/z
700-4000 with external calibration using AB’s 4700 calibration mixture consisting
of the following peptides (in brackets the monoisotopic mass of the singly
protonated ion is given in Da): des-Arg1-bradykinin (904.4681), angiotensin I
(1296.6853), Glu1-fibrinopeptide B (1570.6774) and ACTH 18-39 (2465.1989).
ESI-MS/MS was performed using a quadrupole orthogonal time-of-flight mass
spectrometer (Q-ToF Ultima Global, Micromass/Waters, Manchester, UK)
coupled to a nanoAcquity Ultra Performance LC (Waters, Milford, MA). Peptide
mixtures were trapped and washed on a nanoAcquity column (Symmetry C18, 5
µm, 180 µm x 20 mm) for 3 min using 2 % acetonitrile containing 0.1 % formic
154
acid at 4 µL/min. Peptide fractionation was accomplished using an in-house
packed 12 cm x 75 µm Jupiter C5 or C18 (5 µm; Phenomenex Inc., Torrance, CA)
picofrit column (New Objectives, Woburn, MA). Peptides were eluted using a
binary gradient system consisting of solvent A, 0.1 % formic acid and solvent B,
acetonitrile containing 0.1 % formic acid.
The electrospray ion source was operated in the positive ion mode with a spray
voltage of 3.5 kV. The data-dependent MS/MS mode was used with a 0.6 s survey
scan and 2.4 s MS/MS scans on the three most abundant ion signals in the MS
survey scan, with previously selected m/z values being excluded for 60 s. The
collision energy for MS/MS (25 to 65 eV) was dynamically selected based on the
charge state of the ion selected by the quadrupole analyzer (q1). Mass spectra
were calibrated using fragment ions of Glu1-fibrinopeptide B (MH+ 1570.6774 Da,
monoisotopic mass). In addition, to compensate for temperature drifts, lock mass
correction was used every 30 s for 1 scan using the doubly charged ion of
Glu1-fibrinopepide B ([M+2H]2+ 785,8426 Da, monoisotopic mass).
Mascot software (Matrix Science, London, UK) was used to aid in the
interpretation of tandem mass spectra. Searches were performed using the
SwissProt database with Met oxidation (147.04 Da, monoisotopic mass),
ARP-HNE labeled Cys, His, and Lys (monoisotopic masses 572.2 Da, 606.2 Da,
and 597.2 Da, respectively) as variable peptide modifications. Trypsin/P was
selected as the digesting enzyme allowing for the possibility of one missed
155
cleavage site. Mass tolerances were set to ± 100 ppm for the precursor ion and ±
0.1 Da for the fragment ions. The principal fragment ions were labeled according
to the nomenclature of Biemann [34].
Results and discussion
Labeling
of
a
HNE-modified
Cys-containing
N’-Aminooxymethyl-carbonylhydrazino-D-biotin
and
spectrometric characterization.
peptide
tandem
with
mass
To evaluate the potential of N’-aminooxymethylcarbonylhydrazino D-biotin as a
biotin-labeled aldehyde reactive reagent, we used as a model system the synthetic
peptide Ac-SVVDLTCR-amide. Cysteine sulfhydryl side chains are considered as
likely targets for the electrophile HNE due to their nucleophilic properties.[10]
After reaction with HNE in phosphate buffer (pH 7.4) for 2 hours, the
HNE-conjugated peptide was purified by reverse-phase HPLC and lyophilized.
Mass analysis of the HNE-peptide-conjugate indicated a mass shift by 156 Da
compared to the unmodified peptide confirming that the Michael-type conjugate
was obtained. The oxylipid peptide conjugate was subsequently reacted with ARP
at
room
temperature
for
1
hr.
The
reaction
product
was
isolated
chromatographically. Successful derivatization of the aldehyde functionality by
ARP was confirmed by tandem mass spectrometry.
Tandem mass spectrometric analysis of the ARP-labeled HNE-Cys containing
model peptide, Ac-SVVDLTCR-amide (Pep-HNE-ARP) was achieved by using
156
high energy collision dissociation on a MALDI tandem time-of-flight mass
spectrometer with TOF/TOF optics (Figure 1). The tandem mass spectrum
exhibited an almost complete bn-ion series (b2 to b6). The first C-terminal ion, the
y1 ion, was seen at m/z 174.1. More importantly, we observed the C-terminal
fragment ions, y2* to y5* which retained the ARP-HNE moiety. This ion series
was accompanied by series of ions that indicate neutral loss of the ARP-HNE
moiety from the respective yn* ions (i.e. m/z = 469.3 Th) and these ions are
annotated y4, y3 and y2 in Figure 1. The presence of an Asp-residue in this peptide
causes the relative predominance of the y4* fragment ion. The enhanced cleavage
probability of the peptide amide bond C-terminal to the Asp-residue in this
peptide is particularly pronounced due to the presence of the C-terminal
Arg-residue sequestering the proton [16, 35, 36]. Collectively, the observed
peptide fragment ions confirm that the cysteine residue is modified by a
HNE-ARP moiety. Neutral loss of the HNE-ARP moiety from the precursor ion
MH+ 1402.8 is also evident resulting in the ion at m/z 933.4. The fragment ions
F1 and F2 indicate side-chain fragmentation of the ARP-HNE moiety to some
degree. Noteworthy, the observation of fragment ions that retained the HNE-ARP
moiety are a prerequisite for peptide identification based on the use of
uninterpreted tandem mass spectra acquired under high energy collision
experimental conditions in conjugation with automated database searching.
157
Specificity of ARP towards carbonylated protein modifications.
The oxidation of tryptophan leads to the formation of N-formylkynurenine by
dioxygenation and ring breakage. We tested if the –NH-CHO moiety in the
N-formylkynurenine residue shows reactivity towards ARP. The model peptide
QAKWRLQTL (MH+ 1143.7; Figure 1A in the supplementary material) was
treated with H2O2 overnight and reaction was stopped by adding catalase. The
mass spectrum of the reaction products showed the expected peptide ions that
indicated oxidation of the Trp residue at m/z 1159.7 (+16 Th), m/z 1175.7 (+32 Th)
and m/z 1147.7 (+4 Th) (for structures corresponding to those mass shifts see
Figure 1A in the supplementary material). ARP (20 molar excess) was added to
the oxidized peptide mixture. Mass spectral analysis of the reaction mixture after
treatment with ARP indicated that the –NH-CHO moiety in N-formylkynurenine
is unreactive towards ARP (Figure 1C in the supporting information) due to its
tautomeric stabilization (–NH-CHO and –N=CH(OH)). Similarly, no ARP
reaction
product
was
observed
for
the
decarboxylation
product
of
N-formylkynurenine (structure 3 in Figure 1A in the supporting information).
These results demonstrate that ARP has specificity in its reactivity towards
different forms of protein carbonyl groups.
158
ARP labeling and mass spectrometry of a Michael-type oxylipid protein
conjugate.
In proteins, products of lipid peroxidation, such as the α,β-unsaturated aldehyde
HNE, are able to modify nucleophilic amino acid residues. In particular, Cys and
His residues form readily Michael-type conjugates [10]. In addition, Schiff’s base
formation with the ε-amino group of Lys is also observed and this route leads
potentially to formation of 2-pentylpyrrole [37]. For studying the applicability of
ARP for the efficient labeling of Michael-type oxylipid moieties in proteins we
used E.coli thioredoxin (TRX) as a simple model protein. TRX is composed of
108 amino acid residues, of which 10 are Lys residues. TRX possesses only one
His residue located at position 6. In the oxidized protein, the two cysteine residues,
Cys-32 and Cys-35, are disulfide-linked and, thus, not available for HNE
modification. Using oxidized TRX in our ARP labeling studies enabled us to
focus on the derivatization and characterization of oxylipid-His conjugates and,
accordingly, to complement our Cys-containing peptide model studies.
Adducts of oxidized thioredoxin (TRX) to HNE were obtained by dissolving 1 mg
(82 nMole) thioredoxin in 1 mL 10 mM sodium phosphate buffer (pH 7.4) to
which a 10-fold molar excess of HNE was added. The reaction mixture was
incubated for 2 hr at 37 oC. The stoichiometry of HNE conjugation was checked
by MALDI-MS (Figure 2A). The mass spectral peak at m/z 11,674.9 represents
the unmodified TRX and the ion peak at m/z 11,832.8 corresponds to the addition
159
of one HNE molecule (　m/z 157.9 Th). The mass spectral analysis also indicated
that, although a 10-fold molar excess of HNE to TRX was used in the
modification reaction, a large amount of the TRX remained unmodified. After
confirming the presence of the HNE adduct by MALDI-MS analysis the
HNE-conjugated thioredoxin (TRX-HNE) was reacted with ARP. In the mass
spectrum displayed in Figure 2B it can be seen that the TRX-HNE peak at m/z
11,832.8 is replaced by a peak at m/z 12,145.3 resulting from the ARP-labeled
oxylipid protein conjugate, TRX-HNE-ARP.
A potential problem of approaches that are designed to label the aldehyde
functionality of HNE-peptide or -protein conjugates is the fact that the HNE
conjugate can undergo an intramolecular cyclization reaction in which the
hydroxyl group at position C-4 reacts with C-1 of the aldehyde group to form a
hemiacetal product [9]. Because the hemiacetal no longer contains an
aldehyde-functional group, it is unreactive toward aldehyde-specific reagents and,
therefore, would not be detected using ARP. However, the cyclization equilibrium
is controlled by the pH value. The result from the MALDI-MS analysis shows that
nearly all of the TRX-HNE adduct has been converted to TRX-HNE-ARP which
suggests that the condensation reaction of the aldehyde functionality with ARP is
very efficient under the conditions used, namely 10 mM sodium phosphate, pH
7.4.
160
Tandem mass spectrometric characterization
HNE-modified tryptic peptide T2.
of
the
ARP-labeled
After confirming that the oxylipid-protein conjugate was successfully labeled by
ARP, the unreacted ARP was removed by HPLC. This was followed by digestion
of the sample with trypsin and avidin affinity chromatography to enrich the
ARP-labeled peptides of TRX. The enrichment was performed on the tryptic
peptide mixture rather than the protein so that only the ARP-labeled peptides
would be enriched. The ARP-labeled peptides were isolated from the crude
tryptic digest using an immobilized monomeric avidin column. Avidin is naturally
a tetrameric protein with a very high binding affinity for biotin necessitating very
harsh elution conditions for retrieving biotin-labeled sample. Therefore,
monomeric avidin was used which binds to biotinylated compounds reversibly
and allows for milder elution conditions. The non-labeled peptides were eluted by
washing with PBS buffer. Subsequently, the ARP-labeled peptides were eluted
using 0.3 % formic acid instead of 2 mM biotin in PBS (as described in the
manufacturer’s instruction) in order to facilitate good compatibility with mass
spectrometric analyses. Figure 3 shows the MALDI-MS spectra of the tryptic
digest of TRX-HNE-ARP (Figure 3A) as well as the affinity-enriched peptide
fraction (Figure 3B). Selective enrichment of the peptide with MH+ 2201.2 is
evident
implying
that
this
is
an
ARP-labeled
peptide
(Figure
3B).
MALDI-MS/MS was performed on the peptide ion at m/z 2201.2 to confirm ARP
161
labeling and identify the site of modification (Figure 4).
High energy collision induced dissociation tandem mass spectrometry on the
MALDI ToF/ToF instrument yielded a fragment ion spectrum which identified the
ARP-labeled peptide as the tryptic peptide T2, encompassing the residue 4 to 18
(Figure 4). This peptide contains only one nucleophilic site, namely His-6. Indeed,
the m/z difference of 606.3 Th between the y12 and y13 fragment ions confirmed
the presence of a His-HNE-ARP residue in position 6 (Figure 4). The bn-ions (b3
to b7, b10, b12 and b13) were consistently shifted by 469.3 Th to higher m/z values
compared to the m/z-values that would be expected for an unmodified peptide
with corresponding sequence. The fragment ion at m/z 1731.8 indicates the
neutral loss of the ARP-HNE moiety from the precursor ion MH+ 2201.2. Worth
noting is the ARP-HNE-modified His immonium ion prominently visible at m/z
579.4. The intensive immonium ion formation from HNE-modified His residue
has been also described previously by Bolgar and Gaskell [20] and Fenaille et
al.[19].
We also studied the low energy collision induced dissociation behavior of the
ARP-labeled HNE-modified T2 peptide using a quadrupole time-of flight
instrument equipped with an electrospray ionization source. The triply-charged
molecular ion of the ARP-labeled HNE-modified peptide T2, i.e. the [M+3H]3+
ion at m/z 734.4, was selected as precursor ion (Figure 5). Almost complete yn and
bn fragment ion series were observed. The m/z difference between the b2 ion (at
162
m/z 227) and the b3 ion (at m/z 833) locates the HNE-ARP moiety to the His
residue. Accordingly, the b4 to b12 ions were consistently shifted by 469 Th to
higher m/z values. The modification of the His residue by an ARP-HNE moiety is
also confirmed by the observation of the doubly charged y13 ion at m/z 988 and an
intense His immonium ion at m/z 579.4. Ions that indicate extensive loss and/or
side chain fragmentation of the ARP-HNE moiety were not observed in the
tandem mass spectra acquired under the low energy CID conditions which are
common to qTOF-type instruments.
In order to test the usefulness of the ARP labeling procedure for proteomic-type
applications, we searched the uninterpreted tandem mass spectra from both the
TOF/TOF and the q-TOF instrument against the SwissProt database using the
Mascot software. The HNE-ARP moiety was implemented as variable
modifications on Cys, His and Lys residues. The tandem mass spectrum of the
ARP-labeled T2 peptide acquired on the MALDI-MS/MS instrument obtained an
ion score of 68. Data-dependant analysis on the qToF instrument yielded tandem
mass spectra for the doubly as well as the triply charged ion of the ARP-labeled
T2 peptide; these spectra obtained excellent scores of 78 (data not shown) and 110
(Figure 5), respectively.
Identification of carbonylated proteins in mitochondria.
The ARP labeling approach is currently used in our laboratory for the detection
163
and identification of oxidatively modified proteins in mitochondria isolated from
rat hearts. We repeatedly found several proteins modified by lipid peroxidation
products. For example, the long chain-specific acyl-CoA dehydrogenase,
(ACADL_RAT) was identified as an in vivo target of oxidative modification by
lipid peroxidation products. The tandem mass spectrum of the ARP-labeled
HNE-conjugated peptide (MH+ 2415,09) is depicted in Figure 6. The observed yn
fragment ions (y1, y3-y6, y9, y11, and y12) identify the peptide as the partial
sequence 166-185 of ACADL_RAT. The fragment ion at m/z 2099.3, annotated as
F2, is generated by side chain cleavage at the oxime bond (-O-N=CH-) whereby
the oxylipid moiety is retained at the peptide. In agreement with our previous
observations (see Figures 1 and 4) the fragment ion at m/z 1946.2 arises from
neutral loss of the ARP-HNE moiety (Δm = 468.9 Th). Because the identified
peptide encompasses only a single nucleophilic residue, i.e.Cys-166, the tandem
mass spectral data support the notion that Cys-166 is an in vivo target site for
HNE conjugation. Noteworthy, the same peptide was also found to be modified by
4-oxo-2-nonenal (data not shown). We will communicate these findings in
combination with other biochemical studies in greater detail shortly.
Conclusion
In the above described study, N’-aminooxymethylcarbonylhydrazino D-biotin
(also known as ARP) is introduced as a new reagent for the efficient labeling of
164
carbonyl groups in proteins and peptides. ARP is a hydroxyl amine-functionalized
biotin-containing probe which efficiently reacts with aldehyde/keto groups,
thereby forming an oxime-type (C=N) bond which is sufficiently stable in tandem
mass spectrometric fragmentation studies to allow potentially the localization of
the site of modification. The labeling procedure is a one-step reaction and
proceeds readily under near physiological conditions in phosphate buffer at pH
7.4 at ambient temperature. A potentially important new application for this “old”
probe, which was hitherto used for the detection of abasic sites in DNA, is the
exciting field of proteins as targets of oxidative modifications, for example, by
lipid
peroxidation
products.
The
described
methodology
allows
the
affinity-enrichment of biotinylated oxylipid peptide conjugates and enables their
identification and characterization by mass spectrometry. However, it should be
noted that oxidative modifications to proteins occur via many different
pathways.[2, 4] The chemical diversity of oxidative protein modifications makes
it at best very difficult to use automated searching of uninterpreted tandem mass
spectra as the sole mean of mass spectral data interpretation.
The proof-of-concept study described in this paper introduces ARP as a
potentially useful and alternative probe to hydrazine-based reagents for the
detection and characterization of carbonylated proteins. Studies are currently
ongoing that use the described ARP labeling approach for the detection and
identification of oxidatively modified proteins in cardiac mitochondria.
165
Acknowledgements
Many thanks are due to Duane T. Mooney for his assistance during the early
periods of this project. This work was supported by grants from the NIH/NIA
(AG025372) and Oregon’s Agricultural Research Foundation. The mass
spectrometry core facility of the Environmental Health Sciences Center at Oregon
State University is supported in part by a grant from NIEHS (ES00210).
166
References
[1] T. Finkel, N.J. Holbrook, Nature, 408 (2000) 239.
[2] E.R. Stadtman, Ann. N. Y. Acad. Sci., 928 (2001) 22.
[3] J.D. West, L.J. Marnett, Chem. Res. Toxicol., 19 (2006) 173.
[4] L.J. Marnett, J.N. Riggins, J.D. West, J. Clin. Invest., 111 (2003) 583.
[5] R.L. Levine, E.R. Stadtman, Exp. Gerontol., 36 (2001) 1495.
[6] R.T. Dean, S. Fu, R. Stocker, M.J. Davies, Biochem. J., 324 ( Pt 1) (1997) 1.
[7] R.L. Levine, Free Radic. Biol. Med., 32 (2002) 790.
[8] A. Benedetti, M. Comporti, H. Esterbauer, Biochim. Biophys. Acta, 620
(1980) 281.
[9] H. Esterbauer, R.J. Schaur, H. Zollner, Free Radic. Biol. Med., 11 (1991) 81.
[10] J.A. Doorn, D.R. Petersen, Chem. Res. Toxicol., 15 (2002) 1445.
[11] M.M. Anderson, S.L. Hazen, F.F. Hsu, J.W. Heinecke, J. Clin. Invest., 99
(1997) 424.
[12] L.M. Sayre, M.A. Smith, G. Perry, Curr. Med. Chem., 8 (2001) 721.
[13] J.K. Andersen, Nat. Med., 10 Suppl (2004) S18.
[14] H. Ohkawa, N. Ohishi, K. Yagi, Anal. Biochem., 95 (1979) 351.
[15] R.L. Levine, D. Garland, C.N. Oliver, A. Amici, I. Climent, A.G. Lenz, B.W.
Ahn, S. Shaltiel, E.R. Stadtman, Methods Enzymol., 186 (1990) 464.
[16] T. Reinheckel, S. Korn, S. Mohring, W. Augustin, W. Halangk, L. Schild,
Arch. Biochem. Biophys., 376 (2000) 59.
[17] M.A. Korolainen, G. Goldsteins, I. Alafuzoff, J. Koistinaho, T. Pirttila,
Electrophoresis, 23 (2002) 3428.
167
[18] J.M. Grace, T.L. MacDonald, R.J. Roberts, M. Kinter, Free Radic. Res., 25
(1996) 23.
[19] F. Fenaille, P. Mottier, R.J. Turesky, S. Ali, P.A. Guy, J. Chromatogr. A, 921
(2001) 237.
[20] M.S. Bolgar, S.J. Gaskell, Anal. Chem., 68 (1996) 2325.
[21] B.A. Bruenner, A.D. Jones, J.B. German, Chem. Res. Toxicol., 8 (1995) 552.
[22] A. Musatov, C.A. Carroll, Y.C. Liu, G.I. Henderson, S.T. Weintraub, N.C.
Robinson, Biochemistry, 41 (2002) 8212.
[23] A. Furuhata, T. Ishii, S. Kumazawa, T. Yamada, T. Nakayama, K. Uchida, J.
Biol. Chem., 278 (2003) 48658.
[24] D.L. Carbone, J.A. Doorn, Z. Kiebler, B.R. Ickes, D.R. Petersen, J.
Pharmacol. Exp. Ther., 315 (2005) 8.
[25] D.L. Carbone, J.A. Doorn, Z. Kiebler, B.P. Sampey, D.R. Petersen, Chem.
Res. Toxicol., 17 (2004) 1459.
[26] D.L. Carbone, J.A. Doorn, Z. Kiebler, D.R. Petersen, Chem. Res. Toxicol., 18
(2005) 1324.
[27] B.S. Yoo, F.E. Regnier, Electrophoresis, 25 (2004) 1334.
[28] H. Mirzaei, F. Regnier, Anal. Chem., 77 (2005) 2386.
[29] B.A. Soreghan, F. Yang, S.N. Thomas, J. Hsu, A.J. Yang, Pharm. Res., 20
(2003) 1713.
[30] H. Ide, K. Akamatsu, Y. Kimura, K. Michiue, K. Makino, A. Asaeda, Y.
Takamori, K. Kubo, Biochemistry, 32 (1993) 8276.
[31] K. Kubo, H. Ide, S.S. Wallace, Y.W. Kow, Biochemistry, 31 (1992) 3703.
[32] H. Atamna, I. Cheung, B.N. Ames, Proc. Natl. Acad. Sci. U. S. A., 97 (2000)
686.
[33] J.W. Palmer, B. Tandler, C.L. Hoppel, J. Biol. Chem., 252 (1977) 8731.
168
[34] K. Biemann, Biomed. Environ. Mass Spectrom., 16 (1988) 99.
[35] K.A. Herrmann, V.H. Wysocki, E.R. Vorpagel, J. Am. Soc. Mass Spectrom.,
16 (2005) 1067.
[36] V.H. Wysocki, G. Tsaprailis, L.L. Smith, L.A. Breci, J. Mass Spectrom., 35
(2000) 1399.
[37] R.G. Salomon, K. Kaur, E. Podrez, H.F. Hoff, A.V. Krushinsky, L.M. Sayre,
Chem. Res. Toxicol., 13 (2000) 557.
169
HO
O
C5H11
O
C5H11
HO
S
N
H
S
N
H
O
cysteine Michael-type adduct
(from HNE, +156 u)
O
cysteine hemiacetal
(from HNE, + 156 u)
O
HN
NH
H H
N N
S
O
O
- H2O
NH2
O
N'-Aminooxymethylcarbonylhydrazino D-biotin
(Aldehyde-reactive Probe, ARP, MW 331.3 u)
O
HN
NH
OH
H H
N N
S
O
O
N
C5H11
O
S
N
H
O
ARP derivative of the cysteine Michel-type adduct
(from ARP-HNE, + 469 u)
Scheme 1. Structure and reaction of N’-Aminooxymethylcarbonylhydrazineo
D-biotin (Aldehyde-reactive probe, ARP) for labeling oxylipid–peptide conjugates,
e.g. a HNE-modified cysteine-containing peptide.
170
b2 b3 b4 b5 b6
Ac-S-V-V-D-L-T-C-R-amide
y*5 y*4 y*3 y*2 y1
y4 y3 y2
70
350
630
Mass (m/z)
1086.5
y5*
910
1200
P*
1402.8
y3*
F1
1162.5
y2*
847.4
657.3
b6
746.3
443.2
y4
b5
556.3
328.2
229.1
b3
491.3
10
174.1
y1
b2 y
2
277.1
% Intensity
b4
F2
1075.5
361.2
y3-NH3
899.4
933.4 - HNE-ARP
960.5
100
- HNE-ARP, - H2S
y4*
1480.0
Figure 1. Tandem mass spectrum of the ARP-labeled HNE-peptide conjugate
Ac-SVVDLTCR-amide. Michael-type conjugation of the model peptide was
observed on the Cys-residue. Note, unequivocal localization of the ARP-HNE
moiety to the Cys-residue is obtained by the yn-fragment ions annotated with an
asterisk *.
171
A
11674.9
Doubly-charged
species
5836.7
100
MH+
5915.1
80
70
60
TRX
11832.8
90
% Intensity
MH+
TRX-HNE
50
40
11983.2
30
20
10
4000
6400
8800
11200
13600
16000
O
HN
NH
B
OH
H H
N N
S
O
O
N
O
N
N
11673.5
MH+
90
TRX
TRX
MH+
12145.3
100
5837.6
Doubly-charged
species
70
60
6072.5
% Intensity
80
TRX-HNE-ARP
50
40
30
20
10
4000
6400
8800
11200
Mass (m/z)
13600
16000
Figure 2. MALDI-MS spectra E.coli thioredoxin (TRX) after HNE-modification
(A) and ARP-labeling (B). (A), the difference of ~158 Th between the mass peak
at m/z 11,675 and m/z 11,833 indicates monoadduction by HNE via Michael-type
addition. The low abundant ion at m/z 11,983 may indicate modification of TRX
by a second HNE moiety but via Schiff’s base formation (　m/z 150 Th). (B), the
difference of ~472 Th between the mass peak at m/z 11,674 and m/z 12,145
provides evidence that ARP labeling of the HNE moiety under formation of an
oxime derivative was successful and highly efficient.
172
T4
90
T2 T4
ox
1023.6
20
1267.7
T6
40
2223.1
50
30
2201.1
T2 + HNE-ARP
1731.9
60
1821.9
70
T8
1001.6
% Intensity
80
A
1805.9
100
10
100
90
% Intensity
80
70
60
1240
1680
2120
2560
2201.2
800
T2 (4-18)
IIHLTDDSFDTDVLK
3000
B
HNE-ARP
50
2223.1
40
30
20
10
800
1240
1680
2120
2560
3000
Mass (m/z)
Figure 3. Affinity chromatographic enrichment of the ARP-labeled HNE-peptide
conjugate. (A), MALDI mass spectrum of the unfractionated tryptic digest of
HNE-modified thioredoxin and (B), the avidin affinity-enriched fraction
containing the ARP-labeled HNE-conjugate of peptide T2.
The following non-labeled tryptic peptides were observed: T2,
IIHLTDDSFDTDVLK (1731.9 Da), T4, MIAPILDEIADEYQGK (1806.1 Da),
T6, LNIDQNPGTAPK (1267.4 Da) and T8 GIPTLLLFK (1001.3 Da).
173
* * * *
*
* *
b3 b4 b5 b6 b7
*
b9 b10
*
b12 b13
I-I-H-L-T-D-D-S-F-D-T-D-V-L-K
* y12
y13
10
70
520
1626.8
y12
970
*
b9
1420
b*12
P*
*
y13
1975.0
2056.2
946.6
*
b
5
1047.7
20
748.4
30
263.2
40
1511.7
833.5
b* 4
50
1368.7
b*3
579.4
60
110.1
136.1
% Intensity
70
1162.7
IH*
80
b*10
*
b6
90
1731.8 - HNE-ARP
100
1842.8
1277.7
*
b7
*
b13
1870
2320
Mass (m/z)
Figure 4. MALDI tandem mass spectrometric identification of the ARP-labeled
tryptic peptide T2. Fragment ions marked with an asterisk * retained the
ARP-HNE moiety during high energy collision induced fragmentation. The ion at
m/z 579.4 corresponds to the immonium ion of the ARP-labeled HNE-conjugated
histidine residue. The prominent ion at 1731.8 indicates neutral loss of the
ARP-HNE moiety. This spectrum obtained a Mascot score of 68.
174
*2+
b3
b*3 b4* b5* b6* b*7 b*8 b*9 b*10 b*11 b*12 b*13
I-I-H-L-T-D-D-S-F-D-T-D-V-L-K
y11 y10 y9 y8 y7 y6 y5 y4 y3 y2 y1
2+
y*13
*
IH
575.5
y5
579.4
y6
*
b3
y7
833.7
837.6
690
100
834.7
576.5
838.6
580.4
839.6
835.7
% Intensity
576
578
832 834 836 838 840 842
580
227
b2
y2
260
#
* 2+
b3
417
a2
y3
y4
474
y8
924
y9
a* 3
*
b6
*
b7
1163 1278
y10
805
199
[M+3H]3+ = 734.4 Th
988
* 2+
b7
639
T2 + HNE-ARP
2+
y* 13
*
P
1365
*
b
10
y11
1256
147
y1
0
*
b
8
*
b9
1512
200
400
600
800
1000
1200
1400
1627
*
*
b12
b
11
1730 1843
1600
1800
m/z
Figure 5. ESI tandem mass spectrum of the ARP-labeled tryptic peptide T2. The
triply charged ion at m/z 734.4 was selected for low energy collision induced
fragmentation. Product ions marked with an asterisk * retained the ARP-HNE
moiety. The ion at m/z 579.4 corresponds to the immonium ion of the
ARP-labeled HNE-conjugated histidine residue. The ion marked with # may
represent an internal fragment ion, i.e. [TDD]. This spectrum obtained a Mascot
score of 110.
175
C*IGAIAMTEPGAGSDLQGVR
y5
572.46
ACADL_RAT, aa 166-185
100
Precursor MH+ 2415.09
O
HN
F2
NH
C5H11
H
N
H
N
O
S
O
0
69.0
1946.2
902.6
564.8
y12
1060.6
Mass (m/z)
1556.4
P*
F2
2099.3
- HNE-ARP
C*-
1185.8
y9
1039.7
y6
670.5
687.5
459.4
332.2
86.1
y3 y4
O
~
1056.7
y1
OH
S
O
H2N
175.16
% Intensity
y11
N
2052.2
2548.0
Figure 6. MALDI tandem mass spectrum of the ARP-labeled in vivo
HNE-modified peptide from the long chain-specific acyl-CoA dehydrogenase
(ACADL_RAT, aa 166-185) found in rat heart mitochondria. The fragment ion F2
at m/z 2099.3 indicates side chain fragmentation at the oxime bond whereby the
oxylipid moiety remains bound to the peptide. The ion at m/z 1946.2 indicates
neutral loss of the ARP-HNE moiety. This spectrum obtained a Mascot score of
31.
176
Appendix B
Characterization of 2-alkenal Modified Mitochondrial Proteins by Affinity
Labeling and Mass Spectrometry
Juan Chavez‡, Jianyong Wu‡, William Bisson, Claudia Maier‡*
‡Department of Chemistry, Oregon State University, Corvallis, Oregon 97331
*To whom correspondence should be addressed. Tel: 541-737-9533. Fax:
541-737-2062.
E-mail: claudia.maier@oregonstate.edu
177
ABSTRACT
The modification of proteins by lipid peroxidation products is thought to be an
underlying molecular factor in the pathogenesis of many degenerative diseases
and age related disorders. Here we report on the identification of protein targets
of electophilic 2-alkenals which are formed as a byproduct of naturally occurring
oxidative
stress.
We
employ
the
use
of
N’-aminooxymethylcarbonylhydrazino-D-biotin as an aldehyde reactive probe
(ARP) to label and enrich carbonyl-modified peptides and proteins.
The
enriched ARP labeled peptides are then detected and sequenced using LC-MS/MS
analysis. Employing this method for the analysis of proteins isolated from rat
cardiac mitochondria, 39 unique sites on 27 proteins were identified as being
modified
by
a variety
of
2-alkenal products
including
acrolein, β
-hydroxyacrolein, crotonaldehyde, 4-hydroxy-2-hexenal, 4-hydroxy-2-nonenal
and 4-oxo-2-nonenal.
These sites represent susceptible targets for oxidative
modification and many of the sites are known to be important in the catalytic
function of key mitochondrial enzymes.
Introduction
Since the 1950’s when Harman’s “free radical theory of aging” was introduced
there have been a considerable number of studies focused on understanding the
molecular mechanisms of oxidative stress and it’s impact on living systems [1, 2].
178
It is widely accepted that oxidative stress plays a role in an increasing number of
human health disorders and the aging process [3-9].
Within the rapidly
advancing field of proteomics, several efforts have been devoted to identifying
changes to the proteome in relation to oxidative stress.
Mitochondria are central to studies of oxidative stress, as they are known to be
generators of reactive oxygen species (ROS) within the cell and play a pivotal role
in cell death [7, 8, 10, 11]. There are a number of ROS producing sites within
the mitochondria including complex 1 and complex 3 of the mitochondrial
electron transport chain [12, 13].
Being that the ROS are generated in close
proximity to the mitochondrial membrane, which is rich in polyunsaturated fatty
acids, a whole host of reactive lipid peroxidation products are formed. These
lipid peroxidation products are relatively long lived compared to the free radical
ROS which led to their formation, allowing them to diffuse further from their site
of production to potentially react with proteins and other biomolecules throughout
the cell.
Among the most abundant and studied of these are the highly
electrophilic α,β-unsaturated aldehydes such as 4-hydroxy-2-nonenal (HNE),
4-oxo-2-nonenal (ONE), 4-hydroxy-2-hexenal (HHE), and acrolein [14].
A number of techniques exist for measuring the oxidative modification to proteins
and other biomolecules.
Among the most widely used of these techniques are
assays based on the use of either 2,4–dinitrophenylhydrazine (DNPH) or
thiobarbituric acid (TBARS) derivatization.
While these methods are able to
179
provide researchers with a general overview of levels of oxidative stress induced
modifications, they fail to provide specific information such as the identity of the
modified proteins, and the site and chemical nature of the modification.
In
addition to the traditional colorimetric and immunochemical based methods of
analysis, mass spectrometry has emerged as a powerful technique for studying the
oxidative modification to biomolecules in ageing and disease related research [15].
Towards the goal of identifying the sites of oxidative modification in proteins, a
variety of methods involving chemical derivatization followed by mass
spectrometric detection have been developed. A variety of interesting chemistries
have been employed to label and detect oxidatively modified sites in proteins
including; hydrazide functionalized probes, Girard’s P reagent, and Click
chemistry to name a few [16-18]. One limitation of the many hydrazine-based
reagents for labeling protein carbonyls is the pH-dependant reversibility of the
hydrazone bond and the need for a reduction step to form a stable hydrazine
adduct.
based
We recently demonstrated an alternative approach to the hydrazine
chemistry,
using
N’-aminooxymethycarbonylhydrazino-D-biotin,
a
hydroxylamine functionalized biotin derivative, which forms stable oxime bonds
with protein aldehydes, thus negating the need for a reduction step.
Several recent studies have been devoted to identifying the protein targets and
sites of these lipid-derived electrophiles, in particular targets of HNE [19-21].
Numerous in vitro studies have demonstrated that α,β-unsaturated aldehydes,
180
being electrophilic in nature, covalently modify proteins at nucleophilic sites, in
particular cysteine, histidine and lysine. Cysteine, with its thiol functional group
being the most nucleophilic side chain, and being important to the structure and
function of many enzymes has garnered many studies set out to identify target
cysteine residues in proteins [22-24]. While there has been much success in
identifying protein targets from in vitro studies, there is a void of results when it
comes to the identification of oxidative modifications that occur in vivo. This
information is truly of great importance and interest towards increasing our
understanding of the effects of lipid derived aldehydes on the proteome, and is the
focus of the current study.
Here we expand upon our previously described technique of using
N’-aminooxymethylcarbonylhydrazino-D-biotin as an aldehyde reactive probe
(ARP) which allows for a labeling, affinity enrichment and mass spectrometry
based approach to identify the sites of carbonyl-modification to proteins occuring
naturally within the mitochondria.
ARP is a hydroxylamine functionalized biotin
derivative, which is reactive toward carbonyl-modified proteins, and allows for
affinity enrichment and mass spectrometric detection of the site and chemical
nature of the modification. In our current study we have found this method to be
successful in identifying a number of protein targets with susceptible nucleophilic
sites, which turn out to be modified by a surprising variety of electrophilic
2-alkenal products. We are able to conclusively identify both the site and nature
181
of the modification, thus advancing the study of protein-oxylipid conjugates to a
new level of detail.
Materials and Methods
Chemical Reagents
N’-aminooxymethylcarbonyl hydrazino-D-biotin (Aldehyde Reactive Probe, ARP)
was purchased from Dojindo Molecular Technologies Inc. (Rockville, MD).
Ultralink Monomeric Avidin, Zeba Spin Desalting columns, SuperSignal West
Pico Chemiluminescent substrate were purchased from Thermo Scientific
(Rockford, IL). Neutravidin-HRP was purchased from Pierce. Sequencing grade
modified trypsin was purchased from Promega (Madison, WI).
Ready Gels were purchased from Bio-Rad.
12% Tris-HCl
Immobilon-P PVDF membranes
were purchased from Millipore.
Preparation of rat cardiac mitochondria:
The Institutional Animal Care and Use Committee (IACUC) at Oregon State
University approved the experimental protocol for the use of animals in this study.
Male Fisher 344 rats, obtained from the National Institute of Aging (Bethesda,
MD), were housed in individual plastic cages covered with Hepa filters and
allowed free access to food and water. Rats were anethetized with diethyl ether
and a midlateral incision was made in the chest to remove the heart.
182
Subsarcolemmal mitochondria were isolated from the rat hearts by differential
centrifugation according to the procedures of Palmer with minor modifications by
Suh et al. and stored at -80oC until needed [25, 26]. Mitochondrial samples
containing approximately 1mg total protein were washed twice with 0oC 10mM
NaH2PO4 pH 7.4. The mitochondria were then resuspended in 200 µl of 10mM
NaH2PO4 pH 7.4.
The mitochondria were ruptured by four rounds of
freeze/thaw cycles which involved rapid freezing with liquid N2 followed by
thawing in a cold water bath with 10 minutes of sonication.
Soluble and
membrane proteins were separated by centrifugation at 14,000g for 15min. Pierce
Coomassie Plus protein assay was used to determine the protein concentration.
ARP labeling of Mitochondrial Proteins.
ARP labeling of mitochondrial proteins was accomplished by simultaneously
adjusting the protein concentration to ~1 µg/µl and the ARP concentration to
5mM in 10mM NaH2PO4 buffer pH 7.4. The mixture was then allowed to react
for up to one hour at room temperature.
Excess ARP was removed using Zeba
desalting spin columns (Pierce) with a molecular weight cut off of around 7kDa.
Protein samples were digested with a 1:50 w/w ratio of modified trypsin in
100mM NH4HCO3 pH 8.3 for 6-18 hours at 37oC.
183
SDS-PAGE and Western Blot
One dimension SDS-PAGE was accomplished by loading 20　g of mitochondrial
protein per well of Bio-Rad 12% tris-HCl readymade gels and running them at a
constant 130 V for 60 minutes. For visualization of the protein bands the gels
were incubated with Biosafe Coomassie dye for 1 hr. Western blotting was
accomplished by transferring the proteins from the gel to a PVDF membrane by
applying a constant current of 150 mA for 2 hours. The blot was then blocked
with 5% milk in TBS buffer with 0.1% tween-20 and then washed extensively
before being incubated with 40 ng/mL of HRP-NeutrAvidin for one hour. After
several additional wash steps the gel was exposed to the PicoWest
chemiluminescent substrate for 10 minutes and developed using Kodak BioMax
X-Ray film.
Affinity Enrichment of ARP labeled peptides
100-300 µl of Ultralink monomeric avidin was packed into Handee Mini Spin
Columns, which accommodate solvent addition with a Luer Lok syringe. The
column was washed with 1.5ml of 10mM NaH2PO4, pH 7.4. Irreversible binding
sites, consisting of tetrameric avidin, were blocked by washing with three column
volumes of 2mM D-biotin. To remove excess D-biotin the column was washed
with five column volumes of 2M Glycine-HCl pH 2.8.
The column was then
re-equilibrated by washing with 2mL of 2x phosphate buffered saline (PBS,
184
20mM NaH2PO4 300mM NaCl).
The mitochondrial peptide sample was then
slowly added to the affinity column and the flow thru collected. To remove
non-labeled and non-specifically bound peptides the column was washed with
1mL 2 x PBS followed by 1mL of 10mM NaH2PO4 and finally 1.5mL of 50mM
NH4HCO3 20%CH3OH. The column was then rinsed with 1mL of MilliQ H 2O
before eluting the ARP labeled peptides with 0.4% formic acid 30% acetonitrile.
Collected fractions were then concentrated using vacuum centrifugation and
stored at -20oC before mass spectrometric analysis.
LC-MS/MS Analysis
The resulting ARP enriched peptides were analyzed by LC-MS/MS using a
quadrupole orthogonal time-of-flight mass spectrometer (Q-TOF Ultima Global,
Micromass/ Waters, Manchester, UK) mass spectrometer equipped with a
NanoAcquity UPLC system.
Peptides were fractionated by reverse phase
chromatography on a 100 m i.d x 200mm long BEH column (Waters, Milford,
MA) using a linear gradient of a binary solvent system consisting of solvent A
(2% acetonitrile/ 0.1% formic acid) and solvent B (acetonitrile containing 0.1%
formic acid). The electrospray source was operated in the positive mode with a
spray voltage of 3.5kV.
The mass spectrometer was operated in data dependant
MS/MS mode, performing 0.6 second MS scans followed by 2.4 second MS/MS
scans, on the three most abundant precursor ions in the MS scan with a 60 second
185
dynamic exclusion of previously selected ions. The MS/MS collision energy
(25-65 eV) was dynamically selected based on the charge state of the precursor
ion selected by the quadrupole analyzer.
Mass spectra were calibrated using
fragment ions of Glu1-fibrinopeptide B (MH+ 1570.6774 Da, monoisotopic mass).
Additionally lock spray mass correction was performed on the doubly charged ion
of Glu1-fibrinopeptide ([M+2H]2+ 785.8426 Th) every 30 seconds during the
MS/MS run.
Peptide samples were also analyzed by LC-MALDI-MS/MS by performing
off-line reversed-phase fractionation on a Dionex/LC Packings Ultimate nanoLC
system coupled to a Probot target spotter. Peptides were fractionated over a 75
µm i.d x 150mm C18 PepMap 100 column (LC-Packings) using a binary gradient
consisting of solvent A (5% acetonitrile/ 0.1% trifluoroacetic acid (TFA)) and
solvent B (80% acetonitrile/ 0.1% TFA). A Probot was used to mix the eluting
peptides with a-cyano-4-hydroxycinnamic acid (2mg/mL in 50% acetonitrile
containing 0.1% TFA) and spot the mixture to a 144 spot stainless steel MALDI
target plate.
proteomics
MALDI-MS/MS analysis was performed on an ABI 4700
analyzer
with
TOF/TOF
optics
equipped
with
a
200-Hz
frequency-tripled Nd:YAG laser operating at a wavelength of 355 nm (Applied
Biosystems, Inc., Framingham, MA). Operating in a data dependent reflectron
mode, full MS scans was obtained for the range m/z 700-4000 followed by
MS/MS scans on the 10 most abundant ion signals in the MS scan.
Collision
186
induced dissociation was performed in at a collision energy of 1 keV in a collision
cell with a gas pressure of 6 x 10-7 Torr. Mass spectra were calibrated externally
using the ABI 4700 calibration mixture consisting of the following peptides
des-Arg1-bradykinin ([M+H]+ 904.4681 Da), angiotensin I ([M+H]+ 1296.6853
Da), Glu1-fibrinopeptide B ([M+H]+ 1570.6774 Da), and ACTH 18-39 ([M+H]+
2465.1989 Da).
MS/MS data analysis and identification of ARP-labeled peptides
Tandem mass spectrometric data generated on the Q-TOF were processed into
peak list files using ProteinLynx Global Server v2.3 (PLGS Waters, Manchester,
UK). Both the MS survey and MS/MS scans were calibrated using the lock
mass of m/z 785.8426 with a lock spray tolerance of 0.1Da. MS/MS spectra
were smoothed using the Savitzky-Golay algorithm performing 2 iterations over a
smoothing window of 3 channels.
The survey scan was deisotoped and
centroided using a fast algorithm performing 30 iterations. MS and MS/MS
spectra were centroided with a minimum peak width of 4 channels centroiding the
top 80% with a TOF resolution set at 10,000 and a NP multiplier of 0.7.
MALDI-MS/MS data were processed into peak list files using the Peaks to
Mascot tool in 4000 Series Explorer V3.0 (Applied BIosystems). Default settings
were used except for the peak density for MS/MS spectra was set to a max of 10
187
peaks per 200 Da, the minimum S/N was set to 10 with a minimum peak area of
500 and a max peak/precursor set to 40.
Peak list files of the tandem mass spectrometric data were analyzed with the aid of
Mascot v2.1 (Matrix Science, London, UK) and Peaks Studio v4.5 (Bioinformatic
Solutions Inc., Waterloo, ON Canada) protein database search engines. The Swiss
Prot database v50 (270778 sequences, 99412397 residues) and the following
parameters: taxonomy rodentia (20991 sequences), ± 0.2 Da mass tolerances for
the precursor and fragment ions, possibility of 3 missed proteolytic cleavage sites,
with trypsin/P or semitrypsin selected as the digesting enzyme, and ARP-HHE
(CHK), ARP-HNE (CHK), ARP-Acrolein (CHK), ARP-ONE (CHK), ARP-MDA
(KR), ARP-β-hydroxyacrolein, ARP-crotonaldehyde and methionine oxidation
(M) selected as variable modifications at the residues specified in parenthesis.
Identification of ARP labeled 2-alkenal modified peptides took place in multiple
stages.
First high scoring peptides, those peptides with Mascot scores above the
identity threshold (>50) and/or Peaks scores above 0.90, were considered as
putative identifications.
Secondly intermediate scoring peptides, those with
Mascot scores >20 and/or Peaks scores >0.50 and low scoring peptides, those with
Mascot scores <20 and/or Peaks scores <0.50 were subjected to rigorous manual
inspection before being considered as putative identifications. Manual analysis
included assigning a continuity of b- and y- type ion series, comparison to
homologous peptides with high scores, as well as inspection for the presence of
188
fragment ions not recognized by either Peaks or Mascot, that are specific for the
ARP tag, fragment ions at m/z 227.09 (ARP F1), and m/z 332.14 (ARP), as well
as fragment ions resulting from the neutral loss of all, or part of the
ARP-2-alkenal moiety. The observed neutral loss ions include: (i) loss of 316.1
Da, designated F2, resulting from the fragmentation of the N-O bond within the
structure of ARP generating the neutral loss of an ARP fragment, (ii) loss of
331.14 Da resulting from fragmentation of the oxime bond and loss of the entire
ARP moiety, (iii) loss of the entire ARP-2-alkenal modification by retro-Michael
addition, and (iv) fragmentation of the C-S bond in the side chain of Cys, resulting
in loss of the ARP-2-alkenal-SH neutral species.
All tandem mass spectra from
putatively identified ARP labeled 2-alkenal modified peptides are included in the
supplemental materials with fragment ion annotation.
Calculation of B-factor and Solvent Accessible Area (SAA).
The 3D coordinates of porcine heart mitochondrial type II malate dehydrogenase
(MDH) was retrieved from the crystal structures available in the Protein Data
Bank (Pdb) 1MLD.[27] The B-factor and SAA values for all cysteine residues
of MDH protein in the tetramer conformation were calculated using the program
Molsoft ICM v3.5-1p. The mean B-factor values were derived from the atomic
crystallographic data for each cysteine residue in each of the four monomers of
the tetramer complex of MDH. The reported B-Factor values for each cysteine
189
residue are the average of the four monomers.
The solvent accessible areas were
calculated using a faster modification of the Shrake and Rupley alogorithm
accounting for the presence of hydrogen atoms in the protein structure and using a
water probe radius of 1.4 Å [28].
SAA values were calculated as the average of
a span of three residues, including ±1 residue from the cysteine for each of the
monomers of the tetrameric complex of MDH. The reported SAA values for
each cysteine are the average of the four monomers.
Functional Pathway Network Analysis
The carbonyl-modified proteins identified by LC-MS/MS analysis were evaluated
for their functional biological pathways using the Kyoto Encyclopeda of Genes
and Genomes (KEGG) database.
Cytoscape (v. 2.6.2) was used to construct a
visual representation of the network connections between the modified proteins
and their functional pathways.
Results.
Global Analysis of ARP Labeling of Carbonyl-Modified Mitochondrial
Proteins
Mitochondria are well known sources of ROS and oxidative modification of
mitochondrial proteins is associated with many human health disorders as well as
the aging process [29, 30].
To identify mitochondrial protein targets of
190
electrophilic, 2-alkenal, lipid peroxidation products, we employed a chemical
labeling and affinity enrichment technique using ARP.
The relatively low
abundance of carbonylated proteins under physiological conditions requires the
use of an affinity enrichment approach. The hydroxylamine functionality of ARP
reacts, through a condensation mechanism, with the carbonyl group on the
oxidized protein to form a stable oxime product, adding a mass of 313.1 Da to the
carbonylated protein.
This is a particular advantage of the ARP based approach
over similar affinity labeling methods such as those employing hydrazide
functionalized probes such as DNPH and biotin hydrazide, which require
reduction of the hydrazone bond to form a stable adduct.
The biotin moiety
present in the ARP tag allows for the use of avidin affinity enrichment of
low-level modified peptides, which can then be characterized using LC-MS/MS
analysis. Rat cardiac mitochondria were used for this study based on related
studies indicating that oxidative modification to proteins may play a role in
mitochondrial decay observed in aging and age related diseases [31].
In order to demonstrate the utility of ARP labeling for the detection of oxidatively
modified proteins in rat cardiac mitochondria, we applied a previously
demonstrated method to produce an “ARP-blot” [32].
Mitochondrial protein
samples were reacted with 5mM ARP and then were separated by 1D-SDS-PAGE
followed by Western blotting on a PVDF membrane.
HRP conjugated
NeutrAvidin was used with PicoWest chemiluminescent reagents (Pierce) to
191
visualize the ARP reactive bands (figure 1). Several ARP reactive protein bands
are visible in the Western blot from the mitochondrial sample that was reacted
with ARP.
signal.
The control sample, to which no ARP was added, shows relatively no
Although not conclusively identified, the single HRP-NeutrAvidin
reactive protein band visible in the control is likely due to a naturally biotinylated
protein such as one of the mitochondrial carboxylase enzymes.
Identification of ARP labeled carbonylated proteins by affinity enrichment
and tandem mass spectrometry.
Oxidatively modified mitochondrial proteins were labeled with ARP according to
the previously described procedure. ARP labeled mitochondrial proteins were
tryptically digested and subjected to monomeric avidin affinity enrichment and
LC-MS/MS analysis by ESI-Q-TOF and MALDI-TOF/TOF. The resulting
MS/MS data were searched against the Swiss Prot database with the aid of Mascot
and Peaks software. Database search parameters allowed for the possibility of
the specified ARP labeled post-translational modifications listed in the methods
section. In our approach we used the Mascot and Peaks search results as a guide
for identifying the tandem mass spectra of modified peptides, which were then
confirmed by rigorous manual analysis. Using this approach we identified a
variety of oxidative modifications to peptides originating from several
mitochondrial proteins, the results of which are summarized in table 1. The
192
annotated tandem mass spectra from each of the assigned ARP labeled peptides
are available as supplemental figures S1-S78. In total we identified 39 unique
sites of modification in 27 mitochondrial proteins. The types of modifications
identified consisted of six different 2-alkenal products including; acrolein,
β-hydroxyacrolein, crotonaldehyde, 4-hydroxy-2-hexenal, 4-hydroxy-2-nonenal,
and
4-oxo-2-nonenal.
Three
of
the
modified
peptides
including;
NALANPLYCPDYR from ubiquinol-cytochrome-c reductase core protein 2,
ETECTYFSTPLLLGK
from
malate
dehydrogenase,
and
CIGAIAMTEPGAGDLQGVR from long-chain specific acyl-CoA dehydrogenase,
were identified with more than one type of modification to the same residue.
Several of the proteins identified in table 1 are components of the mitochondrial
electron transport chain (ETC).
Many of these protein subunits have been
previously identified as being carbonylated or modified by HNE in gel-based
studies, which employed Western blot detection using anti-DNP or anti-HNE
antibodies.
For example succinate dehydrogenase [ubiquinone] flavoprotein
subunit, ATP synthase α-chain, ATP synthase β-chain, ATP synthase γ-chain,
cytochrome c oxidase subunit VIb isoform 1, and NADH dehydrogenase
[ubiquinone] flavoprotein 1 were identified with carbonylation and/or HNE
modifications, however the sites and types of modifications were not conclusively
identified [33, 34].
By comparing the results in table 1 with other reports in the literature of studies
193
aimed at identifying sites of oxidative modification to proteins by the exogenous
addition of electrophiles, we can see that there is some overlap of the sites and
proteins identified.
This is particularly interesting in that it suggests that the
modified residues identified in table 1 represent susceptible targets for adduction
by lipid peroxidation products. For example in comparison with previous results
from our group in which we utilized a novel synthetic probe (HICAT) to label and
quantify mitochondrial protein targets of HNE, we identified eight of the same
proteins and sites of modification, including; H 227 of ATP synthase -subunit,
C 100 of ATP synthase D chain, C 141 of ATP synthase O subunit, C 256 of
ADP/ATP translocase 1, C166 of long-chain specific acyl-CoA dehydrogenase, C
93 and C 285 of malate dehydrogenase, and C 191 of ubiquinol-cytochorme-c
reductase complex core protein 2 [18]. In a recent study by Wong and Liebler in
which they added two thiol-reactive electrophilic probes (IAB & BMCC) to
mitochondrial samples from HEK293 cells, they were able to identify 1693 sites
of adduction belonging two 809 proteins.
By comparing their results with the
modified protein sites identified in our study, we can see that there are seven cases
of overlap between the modified residues identified. These sites include C385 of
Acontiase, C256 of ADP/ATP translocase, C100 of ATP synthase D chain, C317
of
Methylmalonate-semialdehyde
dehydrogenase,
C
936
of
NAD(P)
transhydrogenase, C191 of Ubiquinol-cytochrome C reductase complex core
protein 2, and C231 of Voltage-dependant anion selective channel protein 1.
194
The fact that only seven common sites exist out of the 1693 residues identified by
Wong & Liebler and the 39 residues identified in this study raises some interesting
questions as to what actually represent biologically significant targets of
oxidiative stress, and what is the best approach to identify them.
Clearly invitro
studies in which relatively high concentrations of electrophilic probes are added
are capable of identifying a large number of potential target sites, however such
artificial situations may not accurately reflect the most biologically relevant sites.
There is also the question of the site specificity of various electrophiles based on
their differing chemical structures and physicochemical properties. Our results
include a few examples that indicate that despite the variation in size and chemical
structure of these 2-alkenals, their common electrophilic properties allow them to
modify the same nucleophilic sites on proteins. Perhaps the most striking of these
examples is the case of the long chain specific acyl-CoA dehydrogenase, which
was identified with modifications from an unprecedented, five different types of
2-alkenal lipid peroxidation products to Cys166. These modifications included
acrolein, β -hydroxyacrolein, 4-hydroxy-2-hexenal, 4-hydroxy-2-nonenal, and
4-oxo-2-nonenal. Figure 4 displays the MALDI-TOF/TOF generated tandem mass
spectra of the tryptic peptide CIGAIAMTEPGAGSDLQGVR spanning residues
166 to 185 illustrating the diversity of modifications identified on the N-terminal
cysteine.
Several of the sites identidified in this study are known to be very important to the
195
function of the specific proteins.
For example the tryptic peptide including
residue C166 from the long chain specific acyl-CoA dehydrogenase contains
several residues which make up part of a FAD binding site, and adduction of this
site by LPO’s would very likely interfere with the proteins ability to bind FAD
and therefore inhibit it’s ability to catalyze the first step in the -oxidation of
fatty acids.
Another example is cysteine 385 of aconitase, which was identified with an
ARP-acrolein modification.
Aconitase has been shown to be inhibited by
acrolein in a concentration dependent manner [35]. Aconitase is found in the
mitochondrial matrix and is involved in the second step of the citric acid cycle,
catalyzing the reversible isomerization of citrate and isocitrate with cis-aconitate
as
a
reaction
intermediate.
We
identified
the
tryptic
peptide
VGLIGSCTNSSYEDMGR to be modified on C385 by ARP-acrolein. This same
peptide was also identified in a separate study performed in our group in which
(4-iodobutyl)triphenylphosphonium (IBTP) was used as a thiol reactive probe
[24]. Taken together this evidence supports the notion that the susceptibility of
cysteine 385 to modification by electrophilic species could account for the
sensitivity and corresponding loss of activity of aconitase frequently observed
under conditions of oxidative stress [36]. Cysteine 385 is known to be part of a
4Fe-4S iron sulfur cluster which is required for the catalytic activity. A specific
Fe(II) atom, called Fea, is thought to coordinate the OH group of the substrate so
196
as to facilitate it’s elimination. Unlike many iron sulfur clusters this iron-sulfur
cluster is not associated with a redox process.
Instead it has been postulated that
the electronic properties of the 4Fe-4S cluster permit Fea to expand its
coordination shell from the four ligands observed in the x-ray structure of the free
enzyme (three S2- and one OH-) to the six octahedrally arranged ligands observed
in the enzyme substrate complex.
The modification of C385 by a lipid
peroxidation product would undoubtedly inhibit the function of this enzyme, as
irreversible oxidative disruption of the 4Fe-4S cluster is known to cause a loss of
enzymatic activity [37, 38].
The peptide spanning residues 317 to 330 from methylmalonate-semialdehyde
dehydrogenase was identified by both ESI-Q-TOF and MALDI-TOF/TOF
analysis as containing a ARP-acrolein modification to Cys317.
This is
particularly significant in that Cys317 is the nucleophilic acitive site of this
enzyme, which catalyzes the irreversible oxidative decarboxylation of
malonsemialdehyde and methylmalonate semialdehyde to acetyl-CoA and
propionyl-CoA respectively. Obviously such a modification would inhibit the
function of this enzyme.
The ESI-Q-TOF generated MS/MS is shown in figure 2.
Note the presence of fragment ions at 227.09 m/z and 332.14 m/z, which are
generated by the fragmentation of the ARP tag and are useful as diagnostic peaks
to assist in the identification of ARP labeled peptides.
The fact that proteins contain sites susceptible to modification by lipid
197
peroxidation products raises the question of what this means in terms of the folded
structure of the protein molecules.
Why are certain residues targeted over others?
Clearly one would think that in order to be modified, a residue would need to be
located on or near the surface of the protein, and be reasonably exposed to the
surrounding solvent.
A residues susceptibility to modification would also
undoubtedly depend on the side chains pKa value and the chemical
microenvironment in which it is located. We attempted to address this question
for the case of mitochondrial malate dehydrogenase (MDH), which was found
modified by iipid peroxidation products on five cysteine residues (Cys 89, 93, 212,
275, 285). Structural analysis was performed on the crystal structure of porcine
heart (MDH) using the protein modeling software Molsoft ICM V3.5-1p.
Normalized B-Factor values, which serve as a measure of local flexibility of the
protein structure over a given region, were obtained for all eight cysteine residues
in a monomer of MDH in the tetramer conformation. The solvent accessible
areas (SAA) for each of the eight cysteines, which provide a measure of how
exposed each residue is to the surrounding solvent, were also calculated using a
water probe radius of 1.4 Å. Four out of the five cysteine residues we identified as
modified correspond with the cysteine residues with the largest B-Factor values.
Cys 285, which has the largest B-Factor and second largest SAA of all of
cysteines in MDH, was identified being modified by ARP-acrolein and ARP-HHE.
The MALDI-TOF/TOF generated tandem mass spectra of the ARP-acrolein and
198
ARP-HHE modified peptides spanning residues 282-296 are shown below in
figure 3 along with the ribbon structure of MDH illustrating the position of Cys
285 and a table with the corresponding average B-Factor and SAA values for all
cysteine residues in a MDH tetramer. Thus for the case of malate dehydrogenase
it seems that B-factor and SAA values correspond with the susceptibility of
residues towards oxidative modification.
Functional Network Analysis of 2-alkenal Modified Proteins
In order to get an understanding of the potential impact of 2-alkenal modification
to mitochondrial proteins on a systems level we performed a network analysis of
the functional pathways of our identified proteins. The KEGG database was
queried to obtain the functional pathway information and used Cytoscape to
visualize the connections between the proteins and their pathways. The network
of modified proteins is depicted in figure 5.
The network analysis shows that
many of the proteins identified in this study are involved in the functional
pathways of degenerative diseases including Alzheimer’s disease, Huntington’s
disease, and Parkinson’s disease, all in which oxidative stress induced
modification to proteins has been implicated to play an important role.
Many
of the identified proteins are also involved in calcium signaling, which is involved
in mitochondrial regulation of cell death.
199
Discussion:
To the best of our knowledge this work represents the most exhaustive analysis of
native oxidative modifications to mitochondrial proteins by 2-alkenal lipid
derived compounds, through the use of affinity enrichment and mass spectrometry.
In total we identified 27 proteins containing 39 unique nucleophilic sites that were
modified by six varieties of 2-alkenals.
The majority of modified proteins
identified (40%) were members of the respiratory chain. This is to be expected
as the electron transport enzymes are both a major source of ROS within the
mitochondria and in close proximity to a high concentration of poly unsaturated
fatty acids present in the inner mitochondrial membrane.
TCA cycle proteins
comprised the second most abundant category of identified modified proteins at
19%.
A pie chart displaying the percentages of all identified proteins and
modified sites is shown in figure 6A.
Eighty five percent of the 2-alkenal
Michael type additions characterized were present on cysteine residues, while
12% of the modification sites were to histidine with 3% occurring on lysine
(figure 6B).
This is also not surprising as cysteine is well known to be the most
nucleophilic of the three residues.
The proteins and modified sites identified in
this study represent targets within the mitochondrial proteome that are susceptible
to oxidative damage by electrophilic lipid peroxidation products. A few of the
sites identified in this study match with those identified by other studies
employing invitro addition of HNE or other electrophilic probes, followed by
200
affinity enrichment and mass spectrometry.
However there are a number of sites,
which are exclusive to either this study, or the in vitro studies.
In order to get a
better understanding of the role that oxidative modifications play in biological
processes, we feel it is important to characterize the naturally occurring
modifications.
Acrolein modification of Mitochondrial Proteins
The Michael type addition of acrolein to nuleophilic sites in proteins is by far the
most common modification we identified in this experiment, occurring on 26 out
of the 27 carbonylated proteins and comprising a total of 84% of the 2-alkenal
modifications identified (figure 6C).
Acrolein is known to be the most
reactive of the α,β-unsaturated aldehydes and is common environmental toxin
which can be formed from combustion of organic matter or formed biologically
through the metabolism of allyl compounds and lipid peroxidation of
polyunsaturated fatty acids [14, 39].
The toxicity of acrolein has been shown to
be approximately one to three orders of magnitude greater than formaldehyde,
acetaldehyde, and HNE [40]. Exposure to acrolein induces conditions of oxidative
stress by depleting protective nucleophiles such as glutathione (GSH) and
ascorbic acid [41, 42]. Recent studies have shown concentrations of acrolein as
low as 1 µM can induce oxidative stress damage resulting in functional
impairment of mitochondria [35]. Acroleins higher reactivity towards protein
201
nucleophiles could in part explain why we have it detected more frequently than
other α,β-unsaturated aldehydes.
Uchida et al. reported the formation of
NE-(3-formyl-3,4-dehydropiperidino) lysine (FDP-lysine) as a major reaction
product between acrolein and Lys residues. However although the structure of
FDP-lysine contains a free aldehyde group, which would be reactive with ARP,
we were not able to identify any of these types of adducts in our experiments [39].
Being that FDP lysine requires two molar equivalents of acrolein per nucleophilic
residue to form it is likely that FDP-lysine modifications only form under invitro
conditions containing relatively high concentrations of acrolein and are unlikely to
occur under normal physiological conditions.
anti-acrolein
antibodies
used
for
the
It is important to note that most
immunochemical
detection
of
acrolein-protein adducts are specific for FDP-lysine derivatives, so their
sensitivity towards acrolein Micheal type adducts of cysteine residues, such as
those primarily identified in this study, remains unknown.
It is possible that the
ARP approach and the immunological approach are sensitive to two different
populations of acrolein adducts.
Recent studies have shown acrolein to be a
mitochondrial toxin inhibiting electron transport and increasing overall oxidative
stress damage in brain and liver mitochondria [35, 40].
Several important
mitochondrial enzymes including aconitase, ADP/ATP translocase complex I,
complex II, pyruvate dehydrogenase and alpha ketogluterate dehydrogenase have
been shown to be inhibited by acrolein [35, 40].
Out of those six enzymes we
202
were able to identify the sites of acrolein modifications to aconitase, ADP/ATP
translocase, complex I, and complex II in our experiments.
Other Lipid peroxidation products.
In addition to acrolein we also identified a few cases of other lipid peroxidation
products adducted to mitochondrial proteins.
These other products include
4-hyroxy-2-hexenal, 4-hydroxy-2-nonenal, 4-oxo-2-nonenal, crotonaldehyde,
beta-hyroxyacrolein. 4-hydroxy-2-nonenal and 4-oxo-2-nonenal arise primarily
from the lipid peroxidation of the ω-6 fatty acids, linoleic and arachadonic acids,
while 4-Hydroxy-2-hexenal on the other hand, is an end product of the
peroxidation of ω-3 fatty acids, including α-linolenic, eicosapaenoic, and
docosahexanoic acids.
Combined these other 2-alkenal products comprised only
16% of the modifications identified (figure 6. C). It is quite surprising that this
wide variety of LPO’s were all identified as adducting to the same residue sites.
To assess the structural implications of why certain residues are susceptible to
modification we performed structural analysis of malate dehydrogenase by
integrating our mass spectrometry data identifying sites of modification with the
known x-ray crystal structure for MDH. This proved useful in that the sites of
acrolein and HHE adduction correspond to the cysteine residues with high
B-factors and solvent accessible areas. Further studies will be necessary to see if
this correlation extends to other protein systems and if B-factor and SAA values
203
may be useful in predicting susceptible sites of modification by lipid derived
electrophiles.
Evaluation and method validation of the ARP approach (Possible pitfalls &
limitations with ARP approach)
With the rapidly increasing number of new analytical proteomics techniques it is
of critical importance to perform validation of the analytical method to ensure the
results are scientifically sound. Du to the extreme sample complexity, biological
variability and absence of reference standards this becomes a daunting task. We
assessed the robustness of the ARP labeling and enrichment method on several
levels in an attempt to validate our analytical approach.
We tested the
reproducibility of our method by repeating our analysis several times using new
sample preparations from different animals and analyzing the samples on multiple
instruments.
We found that 20 of the ARP labeled oxylipid modified peptides
belonging to 15 proteins were reproducibly identifiable in repeated sample
preparations. The overwhelming presence of acrolein modified peptides towards
other oxylipids raised the question as to whether or not our method was in some
way biased towards acrolein modifications vs. HNE, ONE and other LPO adducts.
To test this hypothesis we added HNE in-vitro into the mitochondria then
followed the same affinity enrichment and mass spectrometric analysis described
above. In this case we identified an overwhelming abundance of HNE modified
204
peptides compared with other modifications. These results demonstrate that our
method is not discriminating against HNE modified peptides, as when HNE is
added, we enrich and detect HNE adducted peptides.
Instead we suggest the
abundance of acrolein modifications is likely due to a relatively high
concentration of acrolein compared to other LPO’s in the mitochondria as well as
acrolein’s higher reactivity towards protein nucleophilic sites.
Additionally many of the modified proteins identified are known to be of
relatively high abundance in the mitochondria.
With such a large dynamic range
of protein concentrations present in the mitochondria, it is extremely difficult to
sample the proteins of lower abundance. This problem becomes exaggerated by
the fact that LPO modified proteins are already of very low abundance compared
to their non-modified variants. It’s possible that this problem could be overcome
by employing a variety of additional fractionaltion techniques, however this
would also require a larger amount of starting material to be sure to have enough
of the lowest abundance modified proteins.
It’s important to consider the possibility that oxidative modifications can occur
during the processing of the sample.
We believe to have minimized
contributions by these artificially generated oxidative modifications by careful
treatment of the sample and by using as gentle of conditions as possible.
However even if such modifications do occur as artifacts of sample preparation
the proteins and amino acid residue sites that were modified still represent
205
susceptible targets to oxidative damage and are therefore still relevant.
The
results shown here are important to increasing our understanding of these types of
modifications to proteins and the biological implications such modifications may
have. The present study is a discovery effort with the goal of identifying the
sites and types of endogenously occurring, 2-alkenal modifications to
mitochondrial proteins.
Now that we have characterized a number of these
adducts, future experiments are needed to determine what impact these
modifications have on the proteins structure and function as well as the function
of the mitochondria and cell as a whole.
Ongoing experiments in our
laboratory include performing quantitative analysis of acrolein, modified peptides
in the mitochondria to better understand how the levels of these modifications
change with ageing.
Additionally in preliminary unpublished results from our
laboratory we have identified several of the same cysteine containing peptide
sequences as being modified by glutathione.
Acknowledgements
This work was supported by grants from the NIH/NIA (AG025372). The mass
spectrometry core facility of the Environmental Health Sciences Center at Oregon
State University is supported in part by a grant from NIEHS (ES00210).
206
References
[1] D. Harman, J. Gerontol., 11 (1956) 298.
[2] D. Harman, Proc. Natl. Acad. Sci. U. S. A., 78 (1981) 7124.
[3] B.S. Berlett, E.R. Stadtman, J. Biol. Chem., 272 (1997) 20313.
[4] E.R. Stadtman, Ann. N. Y. Acad. Sci., 928 (2001) 22.
[5] B.W. Gibson, Int. J. Biochem. Cell Biol., 37 (2005) 927.
[6] M.K. Dennehy, K.A. Richards, G.R. Wernke, Y. Shyr, D.C. Liebler, Chem.
Res. Toxicol., 19 (2006) 20.
[7] J. Sastre, F.V. Pallardo, J. Vina, IUBMB Life, 49 (2000) 427.
[8] N.J. Linford, S.E. Schriner, P.S. Rabinovitch, Cancer Res., 66 (2006) 2497.
[9] E. Cadenas, K.J. Davies, Free Radic. Biol. Med., 29 (2000) 222.
[10] S. Orrenius, V. Gogvadze, B. Zhivotovsky, Annu. Rev. Pharmacol. Toxicol.,
47 (2007) 143.
[11] M. Ott, V. Gogvadze, S. Orrenius, B. Zhivotovsky, Apoptosis, 12 (2007) 913.
[12] A.Y. Andreyev, Y.E. Kushnareva, A.A. Starkov, Biochemistry (Mosc), 70
(2005) 200.
[13] Y. O'Malley, B.D. Fink, N.C. Ross, T.E. Prisinzano, W.I. Sivitz, J. Biol.
Chem., 281 (2006) 39766.
[14] H. Esterbauer, R.J. Schaur, H. Zollner, Free Radic. Biol. Med., 11 (1991) 81.
[15] C. Schoneich, Mass Spectrom. Rev., 24 (2005) 701.
[16] A. Vila, K.A. Tallman, A.T. Jacobs, D.C. Liebler, N.A. Porter, L.J. Marnett,
Chem. Res. Toxicol., 21 (2008) 432.
[17] H. Mirzaei, F. Regnier, J. Chromatogr. A, 1134 (2006) 122.
[18] B. Han, J.F. Stevens, C.S. Maier, Anal. Chem., 79 (2007) 3342.
207
[19] S.G. Codreanu, B. Zhang, S.M. Sobecki, D.D. Billheimer, D.C. Liebler, Mol.
Cell Proteomics, 8 (2009) 670.
[20] P.A. Grimsrud, M.J. Picklo, Sr., T.J. Griffin, D.A. Bernlohr, Mol. Cell
Proteomics, 6 (2007) 624.
[21] D.L. Meany, H. Xie, L.V. Thompson, E.A. Arriaga, T.J. Griffin, Proteomics, 7
(2007) 1150.
[22] H.L. Wong, D.C. Liebler, Chem. Res. Toxicol., 21 (2008) 796.
[23] A. Landar, J.Y. Oh, N.M. Giles, A. Isom, M. Kirk, S. Barnes, V.M.
Darley-Usmar, Free Radic. Biol. Med., 40 (2006) 459.
[24] K. Marley, D.T. Mooney, G. Clark-Scannell, T.T. Tong, J. Watson, T.M.
Hagen, J.F. Stevens, C.S. Maier, J. Proteome Res., 4 (2005) 1403.
[25] J.W. Palmer, B. Tandler, C.L. Hoppel, J. Biol. Chem., 252 (1977) 8731.
[26] B. Paizs, S. Suhai, Mass Spectrom. Rev., 24 (2005) 508.
[27] W.B. Gleason, Z. Fu, J. Birktoft, L. Banaszak, Biochemistry, 33 (1994) 2078.
[28] A. Shrake, J.A. Rupley, J. Mol. Biol., 79 (1973) 351.
[29] E.R. Stadtman, Free Radic. Res., 40 (2006) 1250.
[30] D.C. Wallace, Science, 283 (1999) 1482.
[31]T.M. Hagen, R. Moreau, J.H. Suh, F. Visioli, Ann. N. Y. Acad. Sci., 959 (2002)
491.
[32] W.G. Chung, C.L. Miranda, C.S. Maier, Electrophoresis, 29 (2008) 1317.
[33] K.B. Choksi, W.H. Boylston, J.P. Rabek, W.R. Widger, J. Papaconstantinou,
Biochim. Biophys. Acta, 1688 (2004) 95.
[34] S. Judge, Y.M. Jang, A. Smith, T. Hagen, C. Leeuwenburgh, FASEB J., 19
(2005) 419.
[35] J. Luo, R. Shi, Neurochem. Int., 46 (2005) 243.
208
[36] A.L. Bulteau, M. Ikeda-Saito, L.I. Szweda, Biochemistry, 42 (2003) 14846.
[37] A.L. Bulteau, K.C. Lundberg, M. Ikeda-Saito, G. Isaya, L.I. Szweda, Proc.
Natl. Acad. Sci. U S A., 102 (2005) 5987.
[38] L.V. Matasova, T.N. Popova, Biochemistry (Mosc), 73 (2008) 957.
[39] K. Uchida, M. Kanematsu, Y. Morimitsu, T. Osawa, N. Noguchi, E. Niki, J.
Biol. Chem., 273 (1998) 16058.
[40] L. Sun, C. Luo, J. Long, D. Wei, J. Liu, Mitochondrion, 6 (2006) 136.
[41] N.D. Horton, B.M. Mamiya, J.P. Kehrer, Toxicology, 122 (1997) 111.
[42] J.F. Stevens, C.S. Maier, Mol. Nutr. Food Res., 52 (2008) 7.
209
Table 1. Summary of the mitochondrial proteins modified by lipid peroxidation
derived 2-alkenals and subsequently labeled with ARP and identified by tandem
mass spectrometry. The MS/MS identified peptide sequences containing the
ARP label and 2-alkenal modification from each of the corresponding
carbonylated proteins is shown. The amino acid residue and chemical nature of
the modification are indicated for each case. Biological and functional
assignments were made on the basis of information from the NCBI
(http://www.ncbi.nlm.nih.gov/Pubmed) and the UniProt Knowledgebase
(http://www.uniprot.org) websites.
210
Table 1. (Continued)
211
Table 1. (Continued)
212
Western Blot
-ARP
Control
Mito
+ARP
Coomassie
SDS-PAGE
Mito
-ARP
Control +ARP
STD.
kDa
200
116.3
66.2
45
31
21.5
14.4
6.5
Figure 1. SDS-PAGE and Western blot of mitochondrial proteins from rat hearts.
The difference in signal between –ARP control and the +ARP mitochondrial
samples demonstrates the presence of a large number or ARP reactive proteins, as
well as the selectivity of NeutrAvidin-HRP for the ARP labeled proteins.
213
y13 y12 y11
y10
y8
y7 y6
y5
y4
C*-M-A-L-S-T-A-V-L-V-G-E-A-K
b3
NH
O
S
HN
N
H
N
O
332
O
b8
O
O
HN
NH
N
H
S
227
[M+2H]2+
881.44
100
[ARP+H]+
332.14
[M+H -(ARP-ACR)]+
y10
%
[M+H -ARP]+
974.55
y4
ARP F1
200
675.26 y8
y5 616.38
786.47
503.29
b 2+ 714.43
*8
y2
218.15
100
y7
b*
3
404.22
227.09
0
y6
573.75
300
400
500
600
700
800
1430.75
y12
y11 1158.69
1087.64
y13 1392.78
1289.71
900 1000 1100 1200 1300 1400 1500
m/z
Figure 2. ESI-Q-TOF generated MS/MS of the peptide CMALSTAVLVGEAK,
spanning residues 317-330 of methylmalonate-semialdehyde dehydrogenase
containing an ARP-acrolein modification at C317. C317 also happens to be the
nucleophilic active site for this enzyme. This spectrum also contains the fragment
ions at 332.22 m/z and 227.15 m/z, which are generated from the ARP tag and are
diagnostic of ARP labeling.
214
y12
y11 y10 y9 y8
y7 y6
y4
E-T-E-C*-T-Y-F-S-T-P-L-L-L-G-K
O
NH
b4
b6
S
HN
-LPO-ARP R
b8
N
H
N
O
332
y6
b9
O
O
100
NH
N
H
0
69.0
492.2
y*
12
Cys 285
915.4 Mass (m/z)1338.6
P
1761.8
2185.0
D.
640.55
100
90
0
69.0
939.8
1375.2
Mass (m/z)
[M+H-(HHE-ARP)]+
1702.01
y11
1798.09
1138.81
828.65
741.61
504.4
y10
1239.87
10
y4
y9
975.72
20
b3
y7 y8
430.39
30
332.22
40
ARP F1
[ARP+H]+
60
227.16
% Intensity
70
[M+H-ARP]+
Precursor 2128.99
80
50
Cys 285
F2
y6
B.
Cys 285
[M+H-(ACR-ARP)]+
y10 y
11
*
b*6
b*8 b9
1431.56
b*4 b*5
Cys 285
1702.02
1712.01
1754.98
828.64
y9
1239.85
1330.51
10
y4
1056.67
1096.57
1138.80
20
y7 y8
975.73
50
832.45
60
933.49
[ARP+H]+
[ARP-ACR-SH2]+
332.22
404.24
430.39
% Intensity
70
741.61
F1
227.15 ARP
80
30
C.
S
227
Precursor 2070.95
90
40
O
HN
640.55
A.
b5
1810.6
P
Cys #
B-Factor
SAA (Å2)
89
26.22
24.49
93
26.17
20.55
128
22.32
2.14
132
24.51
22.71
138
19.61
0.22
212
33.71
45.95
275
20.72
2.25
285
35.78
43.84
2246.0
Figure 3. A.MALDI-TOF/TOF generated tandem mass spectrum of the peptide
spanning residues 282-296 from malate dehydrogenase which was found with the
ARP-acrolein and ARP-HHE modifications on Cys 285. Cys 285 is the most
solvent accessible Cys residue according to solvent accessibility analysis. B.
Tandem mass spectrum of the ARP-HHE modified version of the same peptide. C.
Porcine heart MDH tetramer (PDB:1MLD) with the protein backbone displayed
as ribbon (cysteine amino acids colored in yellow) and colored by B-Factor (ICM
v3.6-1d, Molsoft). The residue Cys285 is displayed as cpk and colored by atom
type with the carbon atoms in white. D. Table listing the average B-Factor and
solvent accessible area values data of all Cys residues of MDH protein in tetramer
conformation
215
y18
y15
y14 y13 y12 y11 y10
y9
y8
y7 y6
y5
y4
y3
y2
y1
C*-I-G-A-I-A-M-T-E-P-G-A-G-S-D-L-Q-G-V-R
O
NH
y5
572.46
A.
100
F2
S
HN
-LPO-ARP
N
R
H
N
O
332
O
O
N
H
HN
O
NH
S
227
90
[M+H
–(HHE-ARP)]+
1946.14
2442.0
P
F2
2017
2504
1730.04
1492.8
Mass (m/z)
1967.4
D.
E.
F2
[M+H –(HNE-ARP)]+
C.
y18
P
2050.6
1946.14
1018.2
y13 y14 y15
1417.87
1488.91
10
[M+H –(ONE-ARP)]+
P
1946.05
1967.4
y7 y8 y9 y
543.6
[M+H –(βHA-ARP)]+
B.
y6
2097.35
0
69.0
w5
F2
1946.15
10
y4
1185.80
y2
y12
1286.82
20
y11-NH3
P
2546.0
2442.0
P
F2
2099.27
1056.72
30
y3
687.53
774.57
831.58
902.63
959.66
1039.68
y1
2057.28
40
175.16
50
–(ACR-ARP)]+
60
1946.14 [M+H
1999.23
y11
331.21
332.23 [ARP+H]+
459.38
513.42
70
227.16
274.25
% Intensity
80
2052.2
2548.0
Figure 4. Tandem mass spectra from the peptide spanning residues 166 to 185
from the long chain specific acyl-CoA dehydrogenase which was found to be
modified on Cys 166 by a surprising variey of lipid peroxidation products and
subsequently labeled with ARP. A. Full tandem mass spectrum of the
ARP-acrolein modified version of the peptide. B.,C.,D., and E. display the high
m/z region of the tandem mass spectra for the ARP-　-hydroxyacrolein modified,
ARP-HHE modified, ARP-ONE modified, and ARP-HNE modified versions of
the peptide respectively.
216
Figure 5. Biological pathway analysis from the network of carbonylated
proteins generated with funcation data from the KEGG database. Carbonylated
proteins are colored as yellow nodes and the functional pathways are displayed as
colored nodes. White colored nodes represent pathways for which only one
protein was identified.
217
Figure 6. A. Pie chart displaying the mitochondrial proteins and amino acid
residue sites identified with modifications by lipid peroxidation derived 2-alkenals.
B. Pie chart illustrating the percentage breakdown of the types of residues
identified with Micheal-type additions of lipid peroxidation derived 2-alkenals.
C. Pie chart indicating the percentages of the different types of lipid peroxidation
derived 2-alkenals identified in this study.